Approaches for assessing toxicity of selected contaminants to freshwater mussels

(Bivalvia: Unionidae)

Theodore Walter Valenti Jr.

Thesis submitted to the faculty of the Virginia Polytechnic Institute and State University in partial fulfillment of the requirements

for the degree

Master of Science in the field of

Aquatic Ecotoxicology Department of Biology

Committee members: Dr. Donald Cherry (chair)

Dr. Richard Neves Dr. Rebecca Currie

Defense date:

12-14-04 Blacksburg, Virginia

Copyright 2004.

Key words: freshwater mussel, glochidia, juveniles, toxicity, mercury, chlorine

ii

Approaches for assessing toxicity of selected contaminants to freshwater mussels (Bivalvia: Unionidae)

Theodore Walter Valenti Jr.

Abstract

Laboratory bioassays results suggest that early life stages of freshwater mussels are sensitivity to toxicants. However, toxicological databases for unionids are rather limited because standard test methods are yet developed, and no published studies report endpoints for chronic test that are >9 days. The primary goals of my thesis research were to assess acute and chronic toxicities of chlorine and mercury to early life stages. Inter- and intra-specific species variability in the tolerances of glochidia was observed during acute laboratory bioassays as endpoints were between 8 - 43 µg/L for Hg tests, 1.0 - 2.5 mg/L for NaCl tests, and 70 - 260 µg/L for chlorine (TRC) tests. Glochidia of several species had equal or greater sensitivities to Hg and NaCl than test organisms commonly used to assess environmental risk (i.e. Ceridaphnia dubia, Daphnia magna, Pimephales promelas), whereas they were far more tolerant to TRC than many species. Twenty-one day chronic test endpoints for juveniles were substantially lower than those calculated during acute bioassays with glochidia. Villosa iris 3-mo old juveniles were found to be quite sensitive to Hg as growth was significantly impaired at 8 µg Hg/L. Chronic bioassays with TRC revealed a distinct decrease in susceptibility with increased aged for V. iris (relative sensitivities 3-mo > 6-mo > 12-mo), and that 2-mo old Epioblasma capsaeformis were more sensitive than comparable age classes of V. iris. However, both species were tolerant compared to other aquatic organisms, as the lowest endpoint was 20 µg TRC/L.

iii

Table of contents

Title page.........................................................................................................................................i Abstract.......................................................................................................................................... ii Table of contents .......................................................................................................................... iii List of Tables ................................................................................................................................ vi List of Figures.............................................................................................................................. vii Background ................................................................................................................................ viii Research Objective ..................................................................................................................... xx Structure of Work...................................................................................................................... xxi Work Cited ................................................................................................................................ xxv Acknowledgments ..................................................................................................................... xxx Chapter 1: Sensitivity of mussel glochidia and regulatory test organisms to mercury and a reference toxicant.......................................................................................................................... 1 1.1 Abstract.................................................................................................................................... 2 1.2 Introduction............................................................................................................................. 3

1.2.1 Study goals......................................................................................................................... 6 1.3 Methods.................................................................................................................................... 8

1.3.1 Test organisms ................................................................................................................... 8 1.3.2 Preparation of mercury test solutions ................................................................................ 9 1.3.3 Toxicity tests...................................................................................................................... 9

1.3.3.1 Glochidia..................................................................................................................... 9 1.3.3.2 EPA test organisms ................................................................................................... 10

1.3.4 Reference toxicant tests ................................................................................................... 10 1.3.5 Water chemistry and Hg analysis .................................................................................... 11 1.3.6 Data analysis .................................................................................................................... 12

1.4 Results .................................................................................................................................... 12 1.4.1 Control survivorship ........................................................................................................ 12 1.4.2 Hg salt results................................................................................................................... 12

1.4.2.1 Mercuric chloride...................................................................................................... 12 1.4.2.2 Methyl mercuric chloride.......................................................................................... 13

1.4.3 Reference toxicant results ................................................................................................ 14 1.4.3.1 Glochidia................................................................................................................... 14 1.4.3.2 Standard regulatory test organisms........................................................................... 14

1.4.4. Water chemistry and Hg concentrations ......................................................................... 14 1.5 Discussion............................................................................................................................... 15

1.5.1 Mercury tests.................................................................................................................... 15 1.5.2 Glochidia tolerance compared to standard test organisms............................................... 17 1.5.3 NaCl glochidia reference test........................................................................................... 17

1.6 Acknowledgment................................................................................................................... 19 1.7 References.............................................................................................................................. 20

iv

Chapter 2: The use of sodium chloride as a reference toxicant for evaluating the quality of freshwater mussel glochidia. ...................................................................................................... 31 2.1 Abstract.................................................................................................................................. 32 2.2 Introduction........................................................................................................................... 33

2.2.1 Study Objectives .............................................................................................................. 37 2.3 Materials and Methods......................................................................................................... 37

2.3.1 Test Organisms ................................................................................................................ 37 2.3.2 NaCl Test Design............................................................................................................. 37 2.3.3 Water chemistry ............................................................................................................... 39 2.3.4 Data analysis .................................................................................................................... 39

2.4 Results .................................................................................................................................... 40 2.4.1 Intraspecific variation ...................................................................................................... 40

2.4.1.1 Villosa iris................................................................................................................. 40 2.4.1.2 Epioblasma capsaeformis ......................................................................................... 42 2.4.1.3 Alasmidonta heterodon ............................................................................................. 44 2.4.1.4 Lampsilis fasciola ..................................................................................................... 44

2.4.2 Interspecific variation ...................................................................................................... 45 2.4.2.1 Control Survivorship................................................................................................. 45 2.3.2.2 LC50 values .............................................................................................................. 45 2.4.2.3 Water Chemistry ....................................................................................................... 46

2.5 Discussion............................................................................................................................... 46 2.5.1 Intra-specific variation ..................................................................................................... 46 2.5.2 Inter-specific variation ..................................................................................................... 48 2.5.3 Comparison of LC50 and Control Survivorship.............................................................. 48

2.6 Conclusion ............................................................................................................................. 49 2.7 Acknowledgements ............................................................................................................... 51 2.8 References.............................................................................................................................. 52 Chapter 3: Acute and chronic sensitivity of early life stages of freshwater mussels (Bivalvia: Unionidae) to total residual chlorine relative to water quality criteria of 1984 .. 60 3.1 Abstract.................................................................................................................................. 61 3.2 Introduction........................................................................................................................... 62 3.3 Material and Methods .......................................................................................................... 67

3.3.1 Test organisms ................................................................................................................. 67 3.3.1.1 Glochidia................................................................................................................... 67 3.3.1.2 Juvenile mussel ......................................................................................................... 68

3.3.2 Acute toxicity tests with glochidia................................................................................... 69 3.3.3 Chronic toxicity tests with juveniles................................................................................ 71 3.3.4 Water chemistry ............................................................................................................... 72

3.5 Results .................................................................................................................................... 73 3.5.1 Acute toxicity of TRC to glochidia.................................................................................. 73 3.5.2 Test results for chronic bioassays with V. iris ................................................................. 73 3.5.3 Test results for chronic bioassay with E. capsaeformis ................................................... 74

3.6 Discussion............................................................................................................................... 75

v

3.6.1 Toxicity of TRC to glochidia........................................................................................... 75 3.6.2 Chronic juvenile test ........................................................................................................ 77

3.7 Conclusion ............................................................................................................................. 79 3.8 Acknowledgements ............................................................................................................... 80 3.9 Work Cited ............................................................................................................................ 81 Chapter 4: Acute and chronic toxicity of mercury to early life stages of the rainbow mussel, Villosa iris (Bivalvia: Unionidae) -Short Communication. ........................................ 89 4.1 Abstract.................................................................................................................................. 90 4.2 Introduction........................................................................................................................... 91 4.3 Methods.................................................................................................................................. 93

4.3.1 Test organisms ................................................................................................................. 93 4.3.2 Acute toxicity tests........................................................................................................... 94

4.3.2.1 Glochidia................................................................................................................... 94 4.3.2.2 Juveniles.................................................................................................................... 94

4.3.3 Chronic toxicity test with juveniles ................................................................................. 94 4.3.4 Mercury analysis.............................................................................................................. 95 4.3.5 Statistical analysis............................................................................................................ 96

4.4 Results and Discussion.......................................................................................................... 96 4.4.1 Acute glochidia and juvenile toxicity tests ...................................................................... 96 4.4.2 Chronic 21-day juvenile toxicity test............................................................................... 98

4.5 Conclusions............................................................................................................................ 99 4.6 Acknowledgements ............................................................................................................. 101 4.7 References............................................................................................................................ 102 Curriculum Vitae ...................................................................................................................... 110

vi

List of Tables

Table 1.1. The 24- and 48-h LC50 values, respective 95% confidence intervals, and mean survivorship of glochidia of Lampsilis fasciola, Epioblasma capsaeformis, and Epioblasma brevidens exposed to different concentrations of mercuric chloride (HgCl2). ............................. 23 Table 1.2. The 24- and 48-h LC50 values, respective 95% confidence intervals, and mean survivorship of Ceriodaphnia dubia and Daphnia magna exposed to different concentrations of mercuric chloride (HgCl2)............................................................................................................. 24 Table 1.3. The 24- and 48-h LC50 values, respective 95% confidence intervals, and mean survivorship of glochidia of Epioblasma capsaeformis, Epioblasma brevidens, and Villosa iris exposed to different concentrations of methyl-mercuric chloride (CH3HgCl2)............................ 25 Table 1.4. The 24- and 48-h LC50 values, respective 95% confidence intervals, and mean survivorship of Ceriodaphnia dubia, Daphnia magna, and Pimephales promelas exposed to different concentrations of methyl-mercuric chloride (CH3HgCl2). ............................................ 26 Table 1.5. The 24- and 48-h LC50 values, respective 95% confidence intervals, and percent mortality of glochidia of Lampsilis fasciola, Epioblasma capsaeformis, and Epioblasma brevidens exposed to the reference toxicant NaCl........................................................................ 27 Table 1.6. Mean conductivity, pH, Alkalinity, Hardness, and in- and out-water mercury concentrations for the acute mercury and reference toxicant tests. .............................................. 30 Table 2.1. Summarized test parameters for a 48-h acute reference toxicity test ......................... 54 protocol with freshwater mussel glochidia. Test conditions are based on United States Environmental Protection Agency standard protocol USEPA/600/4-90/027F. ........................... 54 Table 2.2. The LC50 values and mean survivorship for Villosa iris glochidia from eight female mussels in the control, 0.5, 1.0, and 2.0 g NaCl/L treatments. ..................................................... 55 Table 2.3. The LC50 values and mean survivorship for Epioblasma capsaeformis glochidia from five female mussels in the control, 0.5, 1.0, and 2.0 g NaCl/L treatments. .................................. 56 Table 2. 4 The LC50 values and mean survivorship for Alasmidonta heterodon glochidia from two female mussels in the control, 0.5, 1.0, and 2.0 g NaCl/L treatments. .................................. 57 Table 2.5. The LC50 values and mean survivorship for Lampsilis fasciola glochidia from three female mussels in the control, 0.5, 1.0, and 2.0 g NaCl/L treatments. ......................................... 58 Table 3.1. The average viability of glochidia from five species of freshwater mussels exposed to total residual chlorine (TRC). ....................................................................................................... 84 Table 3.2. Comparison of acute toxicological endpoints for common freshwater test organisms, and those for freshwater mussel glochidia generated in our study. .............................................. 85 Table 3.3. Initial length and growth of mussel juveniles exposed to different concentrations of Total Residual Chlorine (TRC) for 21 days.................................................................................. 86 Figure 3.2. Survivorship and proportional growth for 2-, 6-, and 12-mo old Villosa iris exposed to different concentration of TRC................................................................................................. 88 Table 4.1. Summary of test parameters for acute tests with glochidia and juveniles, and the chronic test with juvenile freshwater mussels. ........................................................................... 105 Table 4.2. Mean survivorship and LC50 values after 24, 48, and 72 h for Villosa iris glochidia exposed to different concentrations of mercuric chloride (HgCl2). ............................................ 106 Table 4.3. Mean survivorship and LC50 values after 24, 48, 72, and 96 h for Villosa iris juveniles exposed to different concentrations of mercuric chloride (HgCl2). ............................ 107 Table 4.4. Mean survivorship and growth of 2-mo old juvenile Villosa iris exposed to different concentrations of mercuric chloride for 21-d.............................................................................. 108

vii

List of Figures

Figure 1. Basic life history of freshwater mussels. ...................................................................... xix Figure 1.1. Mean viability of mussel glochidia from Villosa iris, Epioblasma capsaeformis, Epioblasma brevidens, and Lampsilis fasciola in control treatments after 24 and 48 h............... 28 Figure 1.2. The48-h LC50 values of reference toxicant tests conducted with standard regulatory test organisms Ceriodaphnia dubia, Daphnia magna, and Pimphales promelas. Bioassays conducted according to US EPA Protocol (1999) ........................................................................ 29 Figure 2.1. The mean percent control survivorship and LC50 with corresponding standard deviation for glochidia from the mussel species Alasmidonta hererodon, Lampsilis fasciola, Epioblasma capsaeformis, and Villosa iris for a) 24-h test results, and b) 48-h test result.......... 59 Figure 3.1. Diagram of test chamber used for acute glochidia and chronic juvenile toxicity tests. Not drawn to scale. ....................................................................................................................... 87 Figure 4.1 Test design of simulated environment, small scale microcosm used for chronic juvenile mussel toxicity tests. ..................................................................................................... 109

viii

Background

Native freshwater mussels (Mollusca: Bivalvia: Unionidae) are one of the most

imperiled faunas of North America. Currently, less than one quarter of the 297 identified species

have stable populations, whereas 72% of the species are listed as possibly extinct, endangered,

threatened, or of special concern (Williams et al. 1993). Declines are attributed to a number of

stressors, including the introduction of exotic species, overharvest, and habitat disturbance and/or

loss (Neves 1987, Keller and Zam 1991, Goudreau et al. 1993, Jacobson et al. 1993, Williams et

al. 1993, Keller and Ruessler 1997,Yeager et al. 1999a). Several federal environmental policies

issued in the United States since the 1970’s have been essential to the conservation of remaining

mussel assemblages. The Clean Water Act (1977) provided the groundwork for regulating the

discharge of pollution into waterways, and led to substantial improvements in water quality. The

Endangered Species Act (1973) established policies that promoted the conservation of

ecosystems inhabited by threatened and endangered species. The ESA also mandated that

recovery plans be drafted for “listed” species, which spurred extensive research examining the

ecology of these freshwater mussels. Consequently, mussel propagation facilities have been

established, and juveniles from several species have been successfully reared for reintroduction

programs (Ahlstedt 1979, Sheehan et al. 1989, Michaelson and Neves 1995, Jones and Neves

2002). However, despite apparent progress, experts remain concerned about the prolonged

conservation of native freshwater mussels.

By comparing results of historic mussel surveys with those more recent, substantial

declines in taxa richness of freshwater mussels in the United States become evident. More

importantly, accumulation of long-term monitoring data for mussel assemblages has enabled

researchers to make inferences concerning population changes over time (Ahlstedt and Stansbury

ix

1972, Tuberville 1997). One striking trend is the increasing rarity of some species in mussel

assemblages (Sheehan et al. 1989, Diamond et al. 2002). Less apparent, yet more disconcerting,

is the lack of recruitment in assemblages where adults are found (Jacobson et al. 1997, Henley

and Neves 1999). Conservationists are alarmed with current population trends, especially the

lack of young mussels, since it takes several years for most juveniles to reach sexual maturity.

However, since unionids have multiple immature life stages during reproduction (Fig. 1[xix]),

often including a larval parasitic stage that requires a specific fish host, it has proven challenging

for researchers to evaluate the severity to which potential factors are impacting recruitment.

Although in situ studies and surveys are useful for identifying impairment at specific sites, it is

extremely difficult for researchers to isolate variables and distinguish direct cause-effect

relationships because of the complexity of river systems1.

A hypothesis currently proposed by scientists is that recruitment failure is due to early

life stages of freshwater mussels being more sensitive to contaminants than adults. The

aforementioned population trends, and perhaps more decisively laboratory toxicity test results,

provide substantial evidence supporting this hypothesis. Laboratory testing enabled researchers

to assess the relative tolerances of different life stages of unionids (i.e, glochidia, juvenile, adult)

far easier than could be achieved in the field. Furthermore, it provided scientists a better means

to examine cause-effect relationships between impairment and contaminant exposure since

treatment concentrations could be manipulated, while all other variables remained constant.

Perhaps the most comprehensive study to date is Jacobson et al. (1997), which compared the

sensitivity of brooded glochidia (those held with the marsupium of adults), released glochidia

(those released into the water column), encysted glochidia (those attached to host fish),

1 Interaction between multiple non-point and point-source stressors, competitive interaction among organisms, host-fish demographics, varying abiotic and biotic components of ecosystems.

x

transformed juveniles, and adult mussels to copper. This study demonstrated that brooded and

encysted glochidia are very tolerant, and seemingly impervious to copper exposure since they are

sequestered from their external environment by either marsupial or fish host tissue2. Hoggath

and Gaunt (1988) report that encysted glochidia are completely encapsulated by the host tissue

once attached, which may explain their tolerance. Conversely, released Villosa iris glochidia and

juveniles were extremely sensitive to copper, as 24-h LC50 values were 36-80 µg/L and 83 µg/L,

respectively. Adult mussels were found to be more tolerant than either sensitive life stage by

more than 10-fold. During 96 h stream exposures, no adult V. iris mortalities were recorded

despite exposure to copper concentrations as high as 1000 µg/L. Additional studies also report

that early life stages of freshwater mussels are more sensitive than adults (Harrison et al. 1984,

Huebner and Pynnonen 1992), which is typical for most aquatic organisms. Lasee et al. (1991)

reported LC50 values of 141 and 345 µg/L for newly transformed and 14-d-old juvenile

Lampsilis ventricosa exposed to cadmium, respectively, while Naimo et al. (1992) did not

observe any mortalities for adults of the same species exposed to 305 µg Cd/L for 28 d. Keller

and Ruessler (1997) reported no mortality for adults after 96 h for concentrations as high as 350

mg malathion/L, whereas 24- and/or 48-h LC50s for juvenile and glochidia were between 7-374

mg/L.

Studies comparing the acute tolerances of glochidia and juvenile mussels of the same

species are sparse, but of those that do exist, most conclude that glochidia are more sensitive.

Jacobson et al. (1997) examined the copper sensitivities of two species of freshwater mussels. In

their study, acute endpoints for V. iris released glochidia and juvenile mussels were 36-80 µg/L

and 83 µg/L, respectively, while those for Pyganodon grandis were 46-347 µg/L and 33-44 µg/L,

2 Other studies have concluded that brooded glochidia are at risk to contaminant exposure (Huebner and Pynnonen 1992), and this may only apply to heavy metals and/or be species specific.

xi

respectively. Augspurger et al. (2002) reported that glochidia were more sensitive to ammonia

than juveniles despite shorter exposure times. In their study, 24-h L50 values for glochidia of V.

iris, Utterbackia imbecillis, and Actinonaias pectorosa were 3.8, 5.9, and 3.8 mg NH3/L,

respectively, while 96-h LC50 values for juveniles those species were 6.8, 10.6, and 14.0 mg

NH3/L, respectively. Keller and Ruessler (1997) documented that glochidia from several

species, U. imbecillis, V. lienosa, and V. villosa, were more sensitive than juveniles during acute

studies assessing the toxicity of the insecticide malathion. However, additional experiments

assessing the sensitivity of early life stages of the same species are warranted since test methods

employed in former studies were not always consistent, and little is known of intra- and inter-

specific variation in tolerances. Susceptible of different life stages is also probably dependent of

the species and specific toxicant of concern, and statistical interactions between these two

variables are not likely to be uniform.

More importantly, several studies have reported that these early life stages of unionids are

more sensitive to toxicants than test organisms typically used by regulatory agencies to establish

water quality standards and assess environmental risk. In a study comparing Cd, Cr, Cu, Hg, Ni,

and Zn toxicity to U. imbecillis juveniles, and larval stages of Daphnia magna (zooplankton),

Chironomus (midge), and Pimephales promelas (fathead minnow), and Lepomis (bluegill),

Keller and Zam (1991) reported that juvenile mussels were either the most, or second most

sensitive to all the metals except Hg. Furthermore, LC50 values for juveniles were in general

similar to those for D. magna and substantially lower than those for either fish species. Cherry et

al. (2002) conducted acute bioassays with 18 aquatic species to establish an acute site-specific

Criterion Maximum Concentration (CMC) for copper to evaluate the impact of a coal-burning

power plant’s effluent on mussel populations in the Clinch River, VA. Five of the six most

xii

sensitive test organisms were freshwater mussel glochidia, while the other was the mayfly

Isonychia bicolor. Commonly used test organisms, Ceriodaphnia dubia (zooplankton) and P.

promelas, ranked 7th and 15th, respectively. The Genus Mean Acute Values (GMAV) for

glochidia of the four most sensitive mussels species were between 37-60 µg/L, while those for C.

dubia and P. pimephales were 88 and 310 µg/L, respectively. In another study examining

copper toxicity, Jacobson et al. (1993) reported 24-h LC50 values of 83 and 44 µg/L for V. iris

and A. grandis juveniles, respectively, which are again lower than comparable values for C.

dubia and P. pimephales in the Cherry et al. (2002) study. Weinstein and Polk (2001) stated that

glochidia of U. imbecillis are relatively sensitive to photo-activated anthracene compared to

other species (LC50 value=1.93 µg/L), and 6.3 fold more sensitive than oligochaetes to

photoactivated pyrene. Milam and Farris (1998) reported lower mean toxicological endpoints

for C. dubia (1.12 mg ferrous iron/L) and D. magna (2.7 mg Fe/L), than for Leptodea fragilis

glochidia (3.6 mg Fe/L). Pimephales promelas was the most tolerant tested species, as its mean

endpoint was 10.8 mg Fe/L. However, it is important to note that these researchers contrasted 7-

day EC50 values for C. dubia, 48-h LC50 values for D. magna and P. promelas, and 24-h LC50

values for glochidia. Other researchers have generated data that are contrary in that species

typically used in bioassays were more sensitive to contaminants than early life stages of

freshwater mussels (Keller 1993, Masnado et al. 1995); but as with most, these studies only

examined a single species of mussel, U. imbecilis, which is widespread and is assumed to be

tolerant relative to other species.

As adverse effects are observed at low concentrations of toxicant(s) in laboratory studies

with early life stages, researchers have inferred that native freshwater mussels are perhaps one of

the most at-risk aquatic faunas to pollution. However, toxicological data for unionids are not

xiii

included in current water quality standards developed by the United State Environmental

Protection Agency (EPA) for the protection of aquatic life. Regulatory agencies excluded

freshwater mussel toxicity data when calculating water quality criteria (WQC) because they

questioned the validity of test results. Skepticism stemmed primarily from the lack of

standardized test protocols, and quality concerns surrounding the health of test organisms.

Although these concerns are warranted, it seems contradictory for the EPA to exclude freshwater

mussel toxicity data from WQC, while at the same time incorporating results of studies with

other species, such as aquatic insects, that also do not have standardized test methods or quality

assurances established. Regardless, hesitancy to apply laboratory test results to policy makes the

development of standardized test methods for freshwater mussels especially important since

current WQC may not be stringent enough to ensure their protection.

For years, several independent scientific organizations have been involved with research

that has advanced the development of standardized test procedures for early life stage of

freshwater mussels. Johnston et al. drafted a document entitled “Proposed guide for conducting

acute toxicity tests with the early-life stages of freshwater mussels” for the EPA (EPA 1990).

The document played an important role by triggering interaction among laboratories affiliated

with freshwater mussel toxicology and propagation, although it failed to provide sufficient

resolve to many questions pertinent to the development of standardized test methods.

Subsequent studies focused efforts on addressing areas of criticism, and researchers are now

close to drafting a new set of standardized test guidelines for acute bioassays with glochidia

(Wang et al. 2003). Recent efforts show great promise, as experimenters have attempted to

structure test criteria based on protocols established for commonly used aquatic test organisms

xiv

(EPA 1993). The major area of concern still remaining is the lack of QA/QC assurances for

verifying the health and maturity of glochidia used in bioassays.

Juvenile mussels are continually becoming more available for toxicological experiments

because of higher yield from culturing facilities due to improved techniques and greater

knowledge of fish host relationships (Zale and Neves 1982, Michaelson and Neves 1995,

Uthaniwan et al. 2001). As with glochidia bioassays, many components of protocols for acute

laboratory tests with common test organisms are compatible for juvenile mussel experiments

(EPA 1993). Difficulties with obtaining sufficient numbers of individuals for experiments, and

perplexities with culturing some species in the laboratory have slowed development of a

standardized juvenile mussel test protocol. Of the published studies assessing toxicity of

contaminants to juvenile mussels in the laboratory, nearly all include experiments that are

conducted for only short durations (i.e., <96 h), and in most instances, with only one species.

Examples of species typically used are U. imbecillis and V. iris, both of which are relatively

common, adaptive to laboratory conditions, and considered to be quite tolerant compared to

other mussel species.

One of the only published works that examined chronic exposure is “Effect of test

conditions on the toxicity of copper to juvenile unionid mussels” by Keller et al. (1999). The

researchers reported 9-d LC50 values for Lampsilis straminea that were between 11-36 µg Cu/L,

which are substantially lower than comparable values for most other aquatic species. This study

also attempted to determine whether silt, algae, and photoperiod affects toxicity, but reported no

influences, except for the presence of silt that caused a reduction in total Cu in the water column.

However, effects of any of these factors may not have become evident because the period of

exposure was of short duration.

xv

Aquaculture of freshwater mussels has provided toxicologists with valuable information

about living requirements of juveniles; however, it is often not ideal to employ the same

techniques for conducting toxicity experiments. The primary objective of culturing facilities is

to produce a maximum yield of healthy individuals, and consequently, this is often achieved by

incorporating large, high-volume systems. Such designs are not conducive for toxicological

experiments because it is difficult, costly, and space-consuming to maintain multiple replicates

for each treatment. Handling of wastewater from large systems is also troublesome because of

high costs associated with safe disposal due to human health and environmental risks.

Collecting and transporting large volumes of water from impacted river systems for bioassays is

also not practical. The aforementioned limitations have led scientists to begin experimentation

with smaller microcosms; however, determining and incorporating integral components of large-

scale systems to these has proven rather difficult.

Conducting tests in the laboratory for longer durations requires that scientists address

additional components of experimental design, such as feeding, artificial habitat, and test

conditions, that are pertinent to acute tests. Depriving juveniles of food during long-term

experiments will likely cause them to be more susceptible to toxicants. In addition, starvation

will alter toxicokinetics (i.e., absorption, distribution, biotransformation, and excretion of toxic

substances in the body) as filtration rates and metabolic processes (Naimo 1995). Patterson et al.

(1999) observed differences in the glycogen levels in the mantle tissue of adults mussels fed a

controlled diet, versus those starved in quarantine. Irrevocably, feeding juveniles during chronic

tests is pertinent for determining accurate toxicological endpoints, as demonstrated by numerous

studies that established test procedures for other aquatic organisms currently used to assess risk

(EPA 2002). Several researchers have suggested ca. 3 x 107 algae cells/ml as an appropriate

xvi

daily dose for feeding (Gatenby et al. 1997, Yeager et al. 1994); however, concentrations may

fluctuate depending on the specific test designs. In addition, tri-algal suspensions, containing

Chlamydomonas, Ankistrodesmus, and Chlorella, seem ideal (Gatenby et al. 1997), although

single species suspensions of these strains have also been used with success.

Habitat requirements of juvenile mussels, such as the presence of substrate and

aeration/flow, are also important to consider in long-term tests (Gatenby et a, 1997), as their

neglect can ultimately lead to declines in fitness as severe as those for starvation. Hudson and

Isom (1984) found convincing evidence that substrate improves the health and survivorship of

juvenile mussels in laboratory culturing experiments, which seems quite plausible because they

are often found in depositional zones in rivers (Neves and Widlak 1987) and rely more heavily

on pedal feeding than adults (Yeager et al. 1994). Aeration and/or flow in test chambers is also

important to consider when using freshwater mussels as test organisms, as previous studies

document that lotic organisms become stressed in the laboratory experiments if maintained under

static conditions (Kennedy et al. 2003). This is a concern since test organisms in control

treatments must remain healthy throughout the duration of an experiment for results to be valid,

and survivorship must typically be 80% or above in chronic bioassays (EPA 2002). Individuals

will likely alter their normal behavior if habitat requirements are not sufficiently met, which may

affect their susceptibility. Chen et al. (2001) observed that freshwater mussels alter their

behavior (i.e. filtration rate, metabolism) depending on the dissolved oxygen concentration.

Therefore maintaining test organisms in the laboratory during bioassays under conditions that are

similar to those in environmental scenarios will allow researchers to more accurately infer

relative risk in river systems.

xvii

Test conditions, such as temperature, lighting, and water chemistry, can also have

substantial impacts on test results. Temperature extremes can adversely affect organisms and/or

alter respiration and metabolism, which will again influence the susceptibility of an organism to

a particular toxicant. Keller and Ruessler (1997) reported substantial differences in the

sensitivities of juveniles acutely exposed to malathion at temperatures of 25 and 32˚C, though a

similar comparison were less striking in glochidia experiments. Effects of photoperiod is also

important to consider during test design, as recent studies show changes in light intensity and

UV-radiation influence toxicity (Weinstein 2001,Weinstein and Polk 2001); however, behavioral

and physiological effects are not yet well studied. It is critical for researchers to be aware that

environmental conditions affect the steriochemsitry of a toxicant, and only slight variations in

test conditions can have profound effects on toxicity. Furthermore, incorporating uniform test

criteria and conditions will enable researchers to better compare laboratory results from different

locations and times, or with different species and toxicants.

Overall, laboratory toxicity testing with early life stages is vital to conservation of

freshwater mussels because it enables researchers to evaluate risk more accurately. The ability

to not only reduce the complexity, but also manipulate variables in the laboratory that are

uncontrollable in lakes and rivers allow specific stressors to be examined with greater detail. As

a result, scientists are able to more effectively establish cause-effect relationships between

stressors and impairment. Although conducting short-term bioassays in the laboratory are useful

for assessing the toxicity of effluents3 or other highly toxic environmental samples4, it is seldom

that contaminant concentrations reach such elevated levels in aquatic systems. The most

substantial toxicological threat to mussel assemblages, nearly universally among waterways, is

3 Referring to those associated with NYPDS, or equivelant state and local agencies. 4 Examples would included mine drainage, chemical spills, and zones of spray drift.

xviii

the stress induced by chronic, sub-lethal exposure to contaminants. Therefore, conducting long-

term bioassays in the laboratory with juvenile mussels provides a more realistic environmentally

scenario than acute tests. Just as substantial, chronic tests also allow researchers to quantifiably

measure sub-lethal endpoints that are often more sensitive than examining survivorship alone.

Sub-lethal effects are often far more sensitive indicators of impairment than survivorship,

and are especially important to consider for freshwater mussels because of their long life spans.

Foot immobilization (Doherty and Cherry 1988), filtration rate (Doherty and Cherry 1988,

Salanki and V.-Balogh 1989), oxygen consumption (Naimo et al. 1992, Sivaramakrishna et al.

1992), ion levels (Sivaramakrishna et al. 1992) and valve activity (Jacobson et al. 1993) have all

been used as endpoints. However, reduced growth is perhaps a more definitive illustration of

reduced fitness, as juvenile mussels can neither propagate nor survive if they do not grow.

Studies with marine bivalves have observed substantial growth impairments for individuals

exposed to varying levels of heavy metals (Stromgren 1982). Reduced growth also suggests that

individuals are unable to obtain sufficient energy reserves that may lead to latent mortality,

especially for young juveniles if they are unable to develop sufficiently during summer-fall

months to endure winter5.

5 Notably, summer months are typically periods when concentrations of contaminants are highest due to lower flow and less dilution, in addition, metabolic rates are often higher for juveniles because of temperatures are high; therefore risk of toxicity is greatest.

xix

Figure 1. Basic life history of freshwater mussels.

xx

Research Objective

The primary goal of my Master’s thesis (MS) was to assess the toxicity of different

contaminants to early life stages of freshwater mussels (family: Unionidae) under laboratory

conditions. Test designs used in experiments needed to be defendable in court, since

standardized test methods do not yet exist for freshwater mussels. Guidelines for acute glochidia

bioassays were established by integrating methods described in published works; knowledge of

standardized test protocols for commonly used test organisms Ceriodaphnia dubia, Daphnia

magna, and Pimphales promelas; communication with experts in the field; and personal

observation/experimentation. Furthermore, as several researchers have voiced concern for

verifying the health of glochidia used in bioassays, I explored the potential of employing sodium

chloride as a reference toxicant. It was more difficult to develop test procedures for examining

the effect of chronic contaminant exposure to juveniles since little published data exist, and

sufficient numbers of test organisms have only recently become available with improvements in

culturing techniques. Therefore, most of the work was pioneering research, and several pilot

studies to explore different test conditions were conducted before completion of chronic

experiments described in this thesis.

xxi

Structure of Work

There are four distinct chapters presented in this thesis, each was prepared as a

manuscript for submission to journals/books for publication. It is important for readers to realize

that manuscripts in this work may be altered prior to publication in order to address reviewers’

comments or meet a journal’s specific formatting requirements. The decision to structure the

work in this way was made early during my Master’s education, as substantial competition

among laboratories for funding made it important to publish study results. Data collected for

many experiments were conducive for publication because more incorporated pioneer test

designs/concepts for toxicity tests with early life stages of freshwater mussels, and/or addressed

voids in toxicological databases for contaminants.

Chapter 1, entitled “Sensitivity of mussel glochidia and regulatory test organisms to

mercury and a reference toxicant”, was inspired by the results of Cherry et al. (2002), as their

study showed that glochidia from several freshwater mussel species were more sensitive to

copper than commonly used test organisms. However, experiments described in this chapter

examine the sensitivities of glochidia from widespread and threatened mussel species, and

typical test organisms Ceriodaphnia dubia, Daphnia magna, and Pimephales promelas to

inorganic and organic mercury salts. Initially, regulatory agencies were not overly concerned of

the release of Hg because most is emitted as elemental or inorganic forms not highly toxic to

wildlife. However, researchers have since discovered that several abiotic and biotic factors

convert these forms of Hg into methyl-mercury (MeHg), which is 100 times more toxic,

assimilates quickly into biota, and bioaccumulates in food webs. The consumption of

contaminated fish is a substantial Hg exposure risk to humans (French et al. 1999, Mason et al.

2000), and has resulted in the issuing of Fish Consumption Advisories (FCA) in most U.S. states.

xxii

Currently, more than 60% of the FCA issued nationwide are due to Hg (Webber and Haines

2002). Alarm for Hg contamination also has risen because studies suggest that it may be

transported great distances from sources through atmospheric deposition (Slemr et al. 2003,

Walcek et al. 2003), as elevated concentrations have been recorded at remote sites (Huckabee

and Hildebrand 1974, Mason et al. 2000). Furthermore, Hg is highly persistent in aquatic

systems because it may be trapped in the sediment (Cattani et al. 1999, French et al. 1999,

Wiener and Shield 2000), which may be a source of pollution for decades (Wiener and Shield

2000). Several freshwater systems polluted in the 1970’s still have concentrations in the

sediment that are a risk to wildlife (Stansbery 1972, Carter 1977, Cattani et al. 1999, Beckvar et

al. 2000). Since freshwater mussels a threatened aquatic fauna, it is important for researchers to

assess their sensitivity to contaminants of concern. Furthermore, it is of the utmost importance

for researchers to determine the relative sensitivities of early life stages of unionids and

organisms typically used assessing environmental risk if the latter are going to be used as

surrogate test species. Chapter 1 was presented as part of a special platform session at the

Society of Environmental Toxicology and Chemistry (SETAC) 25th annual meeting, Austin, TX,

2002. The data will also be published in the SETAC book “Bivalve Toxicology” (eds. Farris and

Van Hassel). Date of release is unknown. (Peer reviewed by authors, resubmitted, and now

waiting for comments from independent reviewers editing entire book).

Data for Chapter Two was in a manuscript submitted to the journal Environmental

Toxicology and Chemistry (ET&C) entitled “The use of sodium chloride as a reference toxicant

for assessing the health of freshwater mussel glochidia”. However, the work was rejected for

publication because of insufficient evidence supporting the use of NaCl as a reference toxicant,

despite its recommended use by the EPA for other freshwater organisms. However, reviewers

xxiii

agreed that conceptually the approach seems valid. Since it will likely take collaborative effort

among several laboratories to generate sufficient quantitative evidence to verify NaCl as an

acceptable reference toxicant for glochidia, authors on the paper are considering either re-

submitting a revised copy of the version to another journal, or alternatively, presenting it in a

manuscript that focuses on intra-specific variation in tolerances of glochidia from different

adults, as few studies on subject have been published. Concerns were also expressed for

statistical analysis by reviewers, which will be addressed before re-submission of the data.

Jumpin 5.0® was originally used and limitations of the program may have flawed statistical

analysis, therefore SAS will be used.

The third chapter, “Acute and chronic sensitivity of early life stages of freshwater

mussels (Bivalvia: Unionidae) to total residual chorine relative to water quality criteria- 1984”

contains data from acute bioassays with glochidia from several species, and chronic bioassays

with juveniles of two species. Although results for acute bioassays with glochidia are valuable

since several threaten/endangered species were tested, results for chronic tests are far more

intriguing. Three age-classes of V. iris (2-, 6-, and 12-mo old), and one age-class of Epioblasma

capsaeformis (2-mo old) juveniles were exposed to chlorine for 21-d. To our knowledge, these

data are the first to compare the chronic sensitivity of different age-class and species. Chronic

test were conducted in large re-circulating systems that were similar to, at the time of

experiments, the systems used by the Virginia Tech Aquaculture Center to rear juveniles.

Peristalsis pumps fed different concentrations of Cl stock solutions into systems to account for

the high dissipation rate of chlorine so that constant treatment concentrations could be

maintained. An innovative test chamber that enabled researchers to measure growth of an

xxiv

individual juvenile, and allowed flow of test solution through a fine mesh screen, was used

successfully.

Data for Chapter Four, “Acute and chronic toxicity of mercury to early life stages of the

rainbow mussel, Villosa iris (Bivalvia: Unionidae) -Short Communication” were first presented

in a poster at the 25th annual SETAC meeting, Austin, TX, 2002. The principle goals of this

research were to compare the sensitivity of V. iris glochidia and juvenile to acute Hg exposure,

and to determine how chronic Hg exposure affects juvenile survivorship and growth. An

innovative test design that met the unique living requirements of juveniles allowed researchers to

assess the effects of 21-d exposure. Small-scale, aerated microcosms were selected as the test

apparatus because they required small volumes of test solution, which minimized human health

risks, and the production of Hg wastewater. Furthermore, this design improved statistical power

by making it easier to have multiple replicates for different treatments. Tests require only a

small amount of space (i.e., standard incubator, 2 x 3 m flat space), few laboratory requirements

(i.e., air line, temperature controlled environment), and minimal materials that are affordable.

Similar test designs are most applicable for use when attempting to establish standardized

chronic test protocols since bioassays can be conducted at nearly any research facility. Data in

this chapter were originally prepared for submission to ET&C in a full manuscript, but at the

request of the editor, was revised and re-submitted as a Short-Communication. This version has

been accepted by the journal, and is now under re-review with the initial round of peer reviews

completed.

xxv

Work Cited

1) Ahlstedt SA. 1979. Recent mollusk transplants into the North Fork Holston River in southwestern Virginia. Bull Am Malacol Union pp 21- 23.

2) Ahlstedt SA, JD Tuberville. April 1997. Proceedings of a UMRCC symposium- Conservation & management of freshwater mussels II: Initiatives for the future. pp 72-81.

3) Beckvar N, S Salazar, M Salazar, K Finkelstein. 2000. An in situ assessment of mercury contamination in the Sudbury River, Massachusetts, using transplanted freshwater mussels (Elliptio complanata). Can J Fish Aquat Sci 57: 1103-1112.

4) Carter LJ. 1977. Chemical plants leave unexpected legacy for two Virginia Rivers. Science: 198: 1015-1020.

5) Cattani O, D Fabbri, M Salvati, C Trombini, I Vassura. 1999. Biomonitoring of mercury pollution in wetland near Ravenna, Italy by translocated bivalves (Mytilus galloprovincialis). Environ Tox Chem 18: 1801-1805.

6) Chen LY, AG Heath, RJ Neves. 2001. Comparison of oxygen consumption in freshwater mussels (Unionidae) from different habitats during declined dissolved oxygen concentrations. Hydrobiologia 450: 200-214.

7) Cherry DS, JH Van Hassel, JL Farris, DJ Soucek, RJ Neves. 2002. Site-specific derivation of the acute copper criteria for the Clinch River, Virginia. Human Ecol Risk Assess 8: 591-601.

8) Diamond JM, DW Bressler, VB Serveiss. 2002. Assessing relationships between human land uses and the decline of native mussels, fish, and macroinvertebrates in the Clinch and Powell River Watershed, USA. Environ Tox Chem 21: 1147-1155.

9) Doherty FG, Cherry DS. 1988. Tolerance of the Asiatic clam Corbicula sp. to lethal levels of toxic stressors-a review. Environ Pollut 51: 269-313.

10) French KJ, DA Scruton, MR Anderson, DC Schneider. 1999. Influence of physical and chemical characteristics on mercury in aquatic sediment. Water, Air, Soil Pollut 110: 347-362.

11) Gatenby CM, BC Parker, RJ Neves. 1997. Growth and survival of juvenile rainbow mussels, Villosa iris (Lea, 1829) (Bivalvia : Unionidae), reared on algal diets and sediments. Amer Malacol Bull 14: 57-66.

12) Goudrea SE, Neves RJ, Sheehan RJ. 1993. Effects of wastewater treatment plant effluents on freshwater mollusks in the upper Clinch River, Virginia, USA. Hydrobiologia 252: 211-230.

xxvi

13) Hagg WR, ML Warren. 1999. Mantle display of freshwater mussels elicit attack from fish. Freshwater Biol 42: 35-40.

14) Harrison FL, JP Knezovich, DW Rice. 1984. The toxicity of copper to the adult and early life stages of the freshwater clam, Corbicula manilensis. Arch Environ Contam Toxicol 13: 85-92.

15) Henley WF, Neves RJ. 1999. Recovery status of freshwater mussels (Bivalvia: Unionidae) in the North Fork Holston River, Virginia. Am Malacol Bull 15:65-73.

16) Hoggarth MA, AS Gaunt. 1988. Mechanics of glochidial attachment (Mollusca: Bivalvia: Unionidae). J Morph 198: 71-81.

17) Huckabee JW and SG Hildebrand. 1974. Background concentrations of mercury and methylmercury in unpolluted freshwater environments. Barcelona: 6 –10.

18) Hudson RG, BG Isom. 1984. Rearing juveniles of the freshwater mussels (Unionidae) in a laboratory setting. Nautilus 98: 879-883.

19) Huebner JD, KS Pynnonen. 1992. Viability of two species of Anodonta exposed to low pH and selected metals. Can J Zool 70: 2348-2355.

20) Jacobson PJ, RJ Neves, DS Cherry, JL Farris. 1997. Sensitivity of glochidial stages of freshwater mussels (Bivalvia: Unionidae) to copper. Environ Tox Chem 16: 2384-2392.

21) Jacobson PJ, JL Farris, DS Cherry, RJ Neves. 1993. Juvenile freshwater mussel (Bivalvia: Unionidae) responses to acute toxicity testing with copper. Environ Tox Chem 12: 879-883.

22) Jones J, RJ Neves. 2002. Life history and propagation of the endangered fanshell pearlymussel, Cyprogenia stegaria Rafinewque (Bivalvia: Unionidae). J N Am Benthol Soc 21: 76-88.

23) Keller A. 1993. Acute toxicity of several pesticides, organic compounds, and a wasterwater effluent to the freshwater mussel, Anodonta imbecilis, Ceriodaphnia dubia, and Pimephales promelas. Bull Environ Contam Toxicol 51: 696-702.

24) Keller AE, Ruessler DS, Kernaghan NJ. 1999. “Effect of test conditions on the toxicity of copper to juvenile unionid mussels” in Environmental Toxicology and Risk Assessment: Standardization of biomarkers for endocrine disruption, ASTM STP 1364.

25) Keller AE, Ruessler DS. 1997. The toxicity of malathion to unionid mussels: Relationship to expected environmental concentration. Environ Toxicol Chem 16: 1028-1033.

26) Keller AE, Zam SG. 1991. The acute toxicity of selected metals to the freshwater mussel, Anodonta imbecillis. Environ Toxicol Chem 10:539-546.

xxvii

27) Kennedy AJ, Cherry DS, Currie RJ. 2003. Evaluation of ecologically relevant bioassays for a lotic system impacted by a coal-mining effluent using Isonychia biocolor. Environ Monitor Assess 44: 324-331.

28) Masnado RG, SW Geis, WC Sonzogni. 1995. Comparative acute toxicity of a synthetic mine effluent to Ceriodaphnia dubia, Larval fathead minnow, and the freshwater mussel Anodonta imbecilis. Environ Tox Chem 14: 1913-1920.

29) Mason RP, JM Laporte, S Andres. 2000. Factors controlling the bioaccumulation of mercury, methylmercury, arsenic, selenium, and cadmium by freshwater invertebrates and fish. Arch Environ Contam Toxicol 38: 283-297.

30) Michaelson DL, RJ Neves. 1995. Life history and habit of the endangered dwarf wedgemussel Alasmidonta heterodon (Bivalvia: Unionidae). J N Am Benthol Soc 14: 324-330.

31) Milam C, JL Farris. 1998. Risk identification associated with iron-dominated mine discharges and their effect upon freshwater bivalves. Environ Tox Chem 17: 1611-1619.

32) Naimo TJ. 1995. A review of the effects of heavy metals on freshwater mussels. Ecotoxicology 4: 341-362.

33) Naimo TJ, GJ Atchison, LE Hooand-Bartels. 1992. Sublethal effects of cadmium on physiological responses in the pocketbook mussel, Lampsilis ventricosa. Environ Toxicol Chem 11: 1013-1021.

34) Neves, RJ. 1987. Recent die-offs of freshwater mussels in the United States: An overview. Proceedings, Workshop, Die-Offs of Freshwater Mussels in the United States, US Fish and Wildlife and Virginia Polytechnic Institute and State University, Blacksburg, VA, pp. 7-18.

35) Patterson MA, BC Parker, RJ Neves. 1999. Glycogen concentrations in the mantle tissue of freshwater mussels (Bivalvia: Unionidae) during starvation and controlled feeding. Am Malacol Bull 15: 47-50.

36) Salanki J, K, V.-Balogh. 1989. Physiological background for using freshwater mussels in monitoring copper and lead pollution. Hydrobiologia 188/189: 445-454.

37) Sheehan RJ, RJ Neves, HE Kitchel. 1989. Fate of freshwater mussels transplanted to formerly polluted reaches of the Clinch and North Fork Holston Rivers, Virginia. J Fresh Ecol 5: 139-149.

38) Sivaramakrishna B, K Radhakrishnaiah, A Suresh. 1991. Assessment of mercury toxicity by the changes in oxygen consumption and ion levels in the freshwater snail, Pila globosa, and the mussel, Lamellidens marginalis. Bull Environ Contam Toxicol 46: 913-920.

xxviii

39) Slemr F, Brunke EG, Ebinghaus R, Temme C, Munthe J, Waengberg I, Schroeder W, Steffen A, Berg T. 2003. Worldwide trend of atmospheric mercury since 1977. GeoPhys Res Let 30: [np].

40) Stansbery DH. 1972. The mollusk fauna of the North Fork Holston River at Saltville, VA. Bull Amer Malacol Union, Inc. 45-46.

41) Stromgren T. 1982. Effect of heavy metals (Zn, Hg, Cu, Cd, Pb, Ni) on the length growth of Mytilus edulis. Mar Biol 72: 69-72.

42) U.S. Environmental Protection Agency. 2002. Short-term methods for estimation the chronic toxicity of effluents receiving water to freshwater organisms- 4th Ed. EPA/821-R-02-013. Washington, DC.

43) U.S. Environmental Protection Agency. 1993. Methods for measuring the acute toxicity of effluents and receiving waters to freshwater and marine organisms- 4th ed. EPA/600/4-90/027F, Washington, DC.

44) U.S. Environmental Protection Agency. 1990. Proposed guide for conducting acute toxicity tests with the early life stages of freshwater mussels. EPA/68/02/4278. Washington, DC.

45) Uthaiwan K, N Noparatnaraporn, J Machado. 2001. Culture of glochidia of the freshwater pearl mussel Hyriopsis myersiana (Lea, 1856) in artificial media. Aquaculture 195: 61-69.

46) Walcek C, De Santis S, Gentile T. 2003. Preparation of mercury emissions inventory for eastern North America. Environ Pollut 123: 375-381.

47) Wang N, C Ingersoll, E Greer, D White, A Roberts, J Dwyer, T Augspurger, C Kane, C Tibbott, D Neves, C Barnhart. 2003. Developing standardized guidance for conducting toxicity tests with glochidia of freshwater mussels. Presented at Society of Environmental Toxicology and Chemistry, November 9-13, 2003. Austin, TX, USA. Published abstract.

48) Webber HM, TA Haines. 2002. Mercury effects on predator avoidance behavior of a forage fish, golden shiner (Notemigonus crysoleucas). Environ Tox Chem 22: 1556-1561.

49) Weinstien JE. 2001. Characterization of the acute toxicity of photoactivated fluoranthene to glochidia of the freshwater mussel, Utterbacki imbecillis. Environ Tox Chem 20: 412-419.

50) Weinstein JE, KD Polk. 2001. Photoperiod of anthracene and pyrene to glochidia of the freshwater mussel Utterbackia imbecillis. Environ Tox Chem 20: 2021-2028.

51) Williams JD, ML Warren, KS Cummings, JL Harris, RJ Neves. 1993. Conservation status of freshwater mussels of United States and Canada. Fisheries 18: 6-22.

52) Wiener JG, PJ Sheilds. 2000. Mercury in the Sudbury River (Massachusetts, USA): pollution history and synthesis of recent research. Can J Fish Aquat Sci. 57: 1053-1061.

xxix

53) Yeager MM, RJ Neves, DS Cherry. 1999. Competitive interactions between early life stages

of Villosa iris (Bivalvia: Unionidae) and adult Asian clams (Corbicula fluminea). Proceedings of the First Freshwater Mollusk Conservation Society Symposium, pp. 253-259.

54) Yeager MM, DS Cherry, RJ Neves. 1994. Feeding and burrowing behaviors of juvenile rainbow mussels, Villosa iris, (Bivalvia: Unionidae). J N Am Benthol Soc 13: 217-222.

55) Zale AV, RJ Neves. 1992. Fish hosts of four species of lampsiline mussels (Mollusca: Unionidae) in Big Moccasin Creek, Virginia. Can J Zool 60: 2335-2542.

xxx

Acknowledgments I thank my primary advisor, Dr. Donald S. Cherry, for the opportunity to study under him

at Virginia Tech. The wealth of knowledge he taught me from his decades of experience in the

field of aquatic toxicology provided me not only a better understanding of the science and

policies, but also how to best collaborate with regulatory agencies. I also thank my co-advisors

Dr. Richard J. Neves of the U.S. Geological Survey, who drastically improved my writing and

provided invaluable insight about freshwater mussels, and Dr. Rebecca Currie of Roanoke

College’s Biology Department, who aided my development as a research scientist through the

teaching of standard culturing/bioassay protocols. Additional professors also provided me with

substantial guidance throughout my time at Virginia Tech, including Dr. Maury Vallet and Dr.

Madline Schrieber, both of whom were co-authors of a manuscript not in the thesis that

examined impacts of an abandoned arsenic mine on benthic communities. I also owe a debt of

gratitude both to present, and former Cherry Lab students Branden Locke, Matt Simon, Brandi

Echols, Matthew Hull, Alan Kennedy, and Timothy Chad Merricks, Patrick Berry, and Conrad

Peters, in addition to Jake Chaffin of the VT Stream Team, for my educational experience, and

also for some enjoyable times in B-burg. I also greatly appreciated of members of the VT

Aquaculture center, chiefly Rachel Mair, Jess Jones, and Dr. William Henley of the Virginia

Tech. Without the test organisms they provided, my research would not have been feasible. I

would also like to thank John Schmerfeld and Cindy Kane of the U.S. Fish and Wildlife Service

for allowing me to attend meetings discussing declines of freshwater mussels, and test protocols.

But most of all, I want to thank my family, and friends from back home. Without you, my

success would have been more difficult to achieve, and far less meaningful. Mom, dad, sis, and

bro, I love you all very much and will never be able to express my full gratitude.

1

Chapter 1: Sensitivity of mussel glochidia and regulatory test organisms to mercury

and a reference toxicant.

21.1 Abstract

Forty-eight hour acute bioassays were conducted on multiple species of freshwater

glochidia and regulatory organisms to contrast their sensitivities to inorganic and organic

forms of mercury salt. Currently, debate exists on the use of glochidia in bioassays because

of a lack in adequate QA/QC measures. To alleviate this deficiency, I conducted glochidia

NaCl reference tests for some species of freshwater mussels. Results of my study indicated

that the regulatory test organisms Ceriodaphnia dubia and Daphnia magna and glochidia

from Epioblasma capsaeformis, E. brevidens, and Lampsilis fasciola are very sensitive to

mercury (48-h LC50 values ranged from < 4 – 36 µg total Hg/L). Glochidia from Villosa

iris and the fathead minnow, Pimephales promelas, were tolerant to acute mercury

exposure; 48-h LC50 values were 43 and 67 µg total Hg/L, respectively. The studied

clearly showed high levels of inter-specific variation in mercury tolerances and emphasizes

the need for multi-species level risk assessment.

Keywords: glochidia, toxicity, mercury, lethal concentration

31.2 Introduction

Freshwater mussel populations have declined substantially in North America, and

more than two-thirds of the identified species (Unionidae) are classified as extinct,

endangered, threatened, or of special concern (Williams et al. 1993, Naimo 1995, Jacobson

et al. 1997). Although exploitation from commercial overharvest and the introduction of

non-native species have had substantial impacts (Williams et al. 1993, Yeager et al. 1999),

many declines are attributed to anthropogenic stresses that have eliminated or degraded the

natural habitat of mussels (Keller and Zam 1991, Williams et al. 1993, Naimo 1995, Milam

and Farris 1998, Henley and Neves 1999, Diamond et al. 2002, Weinstein 2002). Scientists

have addressed these potential risks by improving agricultural practices, waste

management, and pollution monitoring in the United States, and consequently, water

quality has substantially improved. Furthermore, the implementation of regulatory policies

that are focused on preserving wildlife and the environment, such as the Endangered

Species Act of 1973 and Clean Water Act of 1977, promote to protect not only native

unionids, but also their habitat. However, despite clear progress, there is still concern for

the future conservation of native mussels, as survey efforts have shown little recruitment

(Neves and Widlak 1987, Brenderman and Neves 1993, Henley and Neves 1999).

Researchers have observed that of the remaining diverse mussel assemblages,

many are comprised primarily of older, adult mussels, and few have an abundance of

young mussels present (Henley and Neves 1999, Weinstein 2001). These trends indicate

that populations are unstable and declining. Conservationists are especially concerned

because it may take years for young mussels currently residing in rivers to reach sexual

maturity. The complex life history of unionids has made it difficult for researchers to

determine the causes of reproductive failure. However, there is substantial evidence that

4pollution is a contributing factor, as several laboratory studies have documented that

freshwater mussels, like most aquatic organisms, are more sensitive to contaminants during

their early life stages than as adults (Naimo 1995, Jacobson et al. 1997, Keller and Ruessler

1997, Yeager et al 1999, Weinstein 2001).

Jacobson et al. (1997) conducted a comprehensive study that examined the effects

of copper exposure on various life-stages of freshwater mussels. Their study compared the

sensitivities of Villosa iris glochidia that were brooded (still in the gills of a gravid adult),

released (in the water column), and encysted (attached to fish host). Released glochidia

were impacted at lower copper concentrations (36-80 µg Cu/L) than encysted glochidia

(>400 µg Cu/L). No adverse effects were observed for any treatments in the brooded

glochidia test; however, the highest concentration tested was only 19 µg Cu/L.

Interestingly, released glochidia and juveniles had very similar tolerances, as 24-h LC50

values for glochidia of V. iris and Pyganodon grandis were 36-80 and 46-347 µg Cu/L,

respectively, while those for juveniles were 83 and 44 µg Cu/L. More important, the study

provided clear evidence that early life stages of freshwater mussels have far lower acute

contaminant exposure thresholds than adults, as the 96-h LC50 for adults was >1000 µg

Cu/L.

Only a few other studies have examined the acute tolerances of glochidia and

juvenile mussels of the same species, but most concur with Jacobson et al. (1997) and

report that glochidia are as, or more, sensitive than juveniles in acute exposures. In a study

examining the toxicity of ammonia, Augspurger et al. (2003) recorded higher tolerances for

juveniles than glochidia, despite longer exposure duration. The 96-LC50 values for

juveniles of pheasantshell (Actinonaias pectorosa) and paper pondshell (Utterbackia

imbecillis) were 14.05 and 10.60 mg total ammonia as N/L, respectively, while the

5corresponding 48-h value for glochidia were 3.76 and 5.85 mg total ammonia/L. Similarly,

the mean 96-h LC50 for the rainbow mussel (V. iris) was 6.75 mg total ammonia/L, and the

24-h value for glochidia was 3.79 mg total ammonia /L. Keller and Ruessler (1997)

examined the toxicity of malathion to early life stages of U. imbecillis, little spectaclecase

(V. lienosa), and downy rainbow mussel (V. villosa), and also recorded substantially lower

tolerances for glochidia than for juveniles.

Additional studies have also documented that glochidia are more acutely sensitive

to contaminants than standard regulatory organisms used for Whole Effluent Toxicity

(WET) testing, and US Environmental Protection Agency (US EPA) Water Quality Criteria

(WQC). Cherry et al. (2002) compared the acute sensitivities of 17 species of freshwater

organisms to copper. Four of the five most sensitive test organisms were freshwater mussel

glochidia, while standard regulatory test organisms Ceriodaphnia dubia and Pimephales

promelas ranked 6th (88µg Cu/L), and 14th (310 µg Cu/L), respectively. The Genus Mean

Acute Values (GMAV) for glochidia of the four most sensitive mussels species ranged

from 37 to 60 µg Cu/L. Studies that examined the toxicity of ammonia to early life stages

of freshwater mussels also reported LC50 values that are within the ranges described for

standard test organisms C. dubia, P. promelas, D. magna, and Oncorhynchus mykiss

(rainbow trout) (Mummert et al. 2003, Goudreau et al. 1993). Milam and Farris (1998)

noted that glochidia of Leptodea fragilis were more sensitive than P. promelas to partially

treated mine water, but less sensitive than D. magna and C. dubia. However, their study

contrasted 24-h acute glochidia LC50s with 48-h acute LC50s for D. magna and 7-day

fecundity EC50s for C. dubia. Although the results of the aforementioned studies may

influence freshwater regulatory policy, agencies are hesitant to accept test results because

there is concern for the effectiveness of glochidia as test organisms in the laboratory.

6 Guidelines for conducting acute toxicity tests with early life stages of freshwater

mussels were drafted in 1990 (US EPA 1990). The effort brought laboratory toxicity

testing with freshwater mussels to the foreground of aquatic toxicology, but failed to

address several aspects essential for the development of a standard protocol. The primary

criticism was the use of glochidia in toxicity tests that were obtained from gravid adults

collected from rivers. There is concern that environmental variables, such as pollution or

nutrient availability, may affect the ability of gravid females to produce fit offspring. The

maturity of glochidia collected from different adults of the same species will likely vary, as

not all individuals from a species have synchronized reproductive cycles. The time of

season that mussels are obtained from the field may also influence maturity of glochidia, as

unionids can be categorized into long- and short-term brooders (Jacobson et al 1997).

Unhealthy or immature glochidia are likely to be more susceptible to contaminant exposure

(Huebner and Pynnonen 1992, Goudreau et al. 1993, Jacobson et al. 1997), and their use in

tests may lead to biased, false-positive results. Although verifying test organism health is a

universal concern for all toxicological studies, it is especially problematic for research with

glochidia because researchers are still unsure of appropriate methods. There have been

substantial strides towards establishing acceptable test parameters and methodologies for

glochidia tests, but efforts will go unwarranted unless better techniques for assessing the

health of glochidia are developed.

1.2.1 Study goals

The primary objective of this study was to compare the sensitivities of glochidia of

different species of freshwater mussels to Hg by conducting laboratory tests with organic

and inorganic Hg salts. Many freshwater systems are contaminanted by Hg pollution, as

anthropogenic sources, such as the incineration of medical wastes, disposal of Hg-laden

7material, industrial processing, pesticide use, and the burning of fossil fuels, have made it

more available in ecosystems. Although most Hg is emitted in elemental or inorganic

forms that are not highly toxic, several abiotic and biotic factors may facilitate the

conversion of these forms into methyl-mercury (MeHg) in water (Barkay et al. 1997,

Wiener and Shields 2000, Mauro et al. 2002). This organic form of Hg is highly toxic to

aquatic life, and has been documented to bio-accumulate in food webs (Barkay et al. 1997,

French et al. 1999, Mason et al. 2000, Wiener and Shields 2000, Mauro et al. 2002). The

US EPA is currently reassessing the WQC for Hg, as researchers have become more aware

of the threat it poses to humans and wildlife (Moore et al. 2003). Fish Consumption

Advisories (FCA) for Hg have been issued in nearly every U.S. state (French et al. 1999,

Mason et al. 2000, Webber and Haines 2003). However, recent studies examining the

sensitivities of freshwater organisms are sparse, and results from earlier studies may be

flawed because technology for measuring low concentration of Hg did not exist.

Furthermore, there is little known about the sensitivity of freshwater mussels to Hg, despite

documented declines in polluted (Henley and Neves 1999, Beckvar et al. 2000). It is

pertinent to address these voids because a more comprehensive species database will be

needed to establish appropriate water standards.

Another objective of this study was to compare the Hg sensitivities of glochidia to

those of standard regulatory organisms, C. dubia, D. magna, and P. promelas. Several

studies have noted that glochidia are extremely sensitive compared to the larvae stages of

other aquatic biota (Jacobson et al. 1997, Weinstein 2001, Weinstein and Polk 2001,

Cherry et al. 2002). I wanted to determine whether standard, freshwater regulatory test

organisms are suitable surrogate test organisms for assessing Hg exposure risks to

glochidia. Environmental risk is often inferred by conducting toxicity tests with standard

8monitoring organisms that are sensitive to most toxicants. This approach should not be

implemented for assessing risk to freshwater mussels until the relative tolerances of the

respective genera are discerned.

The final objective of this study was to expose glochidia to sodium chloride (NaCl)

to determine whether it is an appropriate reference toxicant. Tests were conducted based

on methods described in protocol for standard freshwater test organisms (US EPA 1993).

Reference toxicity test measures are useful QA/QC assurances for standard test organisms

because they enable researchers to evaluate the relative health of the test organisms, verify

the acceptability of test conditions or procedures, and validate toxicity tests results.

Reference toxicant tests are supposed to be conducted monthly at culturing facilities, and

concurrently with acute and chronic WET testing with standard test organisms. Similar

approaches have not been applied to glochidia, and the inadequacy of current methods for

assessing the health of glochidia must be addressed for regulatory agencies to be willing to

incorporate test results into environmental policy.

1.3 Methods

1.3.1 Test organisms

Gravid specimens of Lampsilis fasciola (wavyrayed lampmussel), V. iris (rainbow

mussel), Epioblasma capsaeformis (oyster mussel), and E. brevidens (cumberland

combshell) were obtained from the Virginia Polytechnic Institute & State University

(VPI&SU) Aquaculture Center in Blacksburg, VA. Gravid adults of the various species

were collected from the Clinch River, VA, and stored at the Aquatic Wildlife Conservation

Center in Marion, VA. Adult mussels were acclimated to laboratory conditions for at least

48 h before the glochidia were extracted. Glochidia were removed by gently prying open

the valves of a gravid female, puncturing the gill tissue with a sterile, water-filled syringe,

9and then injecting water to flush the glochidia. They were loaded into test chambers < 2 h

after extraction.

Daphnids, C. dubia and D. magna (< 24 h-old), were cultured at the VPI&SU

Aquatic Toxicology Laboratory according to standard procedure (APHA 1998). Organisms

were cultured in a 80:20 mixture of moderately hard, synthetic water (EPA100) (US EPA

1993) and filtered reference water at 25+1° C under 16:8 light: dark photoperiod, and were

fed a diet of unicellular algae (Selenastrum capricornutum) and YCT (Yeast-Cereal leaves-

Trout chow). Fathead minnows were obtained from a commercial supplier (Aquatox, Inc,

Hot Springs, AR).

1.3.2 Preparation of mercury test solutions

Mercuric chloride (MC) and methyl-mercuric chloride (MMC) salts were used to

create the inorganic and organic test solutions, respectively. Test concentrations were 8,

15, 30, 60, and 120 µg/L total Hg, plus a control, in all bioassays, except for some C. dubia

and D. magna tests when the highest concentration, 120 µg/L, was replaced with the lower

concentration of 4 µg/L total Hg.

1.3.3 Toxicity tests

1.3.3.1 Glochidia

Because a protocol has yet to be established for glochidia bioassays, we attempted

to adhere to the test design described in US EPA protocol (1993) for standard freshwater

test organisms. The main modification was an increase in the number of test organisms per

replicate. The small size of glochidia makes them difficult to monitor individually, and

therefore, I assessed viability for a sub-sample of individuals from each replicate. This

approach provided a more accurate estimate of viability per replicate, and also minimized

problems from potential handling stress.

10 Glochidia were randomly distributed to 50-ml glass beakers filled with ~35 ml of

test solution. There were eight replicates of 50 –100 glochidia for each treatment.

Viability was assessed in four randomly selected replicates after 24 h, and the remaining

four replicates were assessed after 48 h. Tests were conducted at 20+1° C, under a 12:12

light:dark photoperiod.

Glochidia viability was assessed through a NaCl response test, similar to that

described by Goudreau et al. (1993) and Huebner and Pynnonen (1992). A sample of

glochidia from a replicate was transferred with a fine-tip glass to a petri dish for

observation using a dissecting scope. The total number of open and closed glochidia was

recorded, after which a concentrated NaCl solution was added. Any glochidia closed prior

to, or remaining open after the addition of the salt solution were recorded as functionally

dead.

1.3.3.2 EPA test organisms

Acute 48-h toxicity tests were conducted with C. dubia, D. magna, and P. promelas

according to US EPA standard protocol (1993). Cladoceran bioassays were conducted in

50-ml glass beakers with approximately 35 ml of test solution. There were 4 replicates of 5

individuals for each treatment. Pimephales bioassays were conducted in 300-ml glass

beakers filled with ~250 ml of test solution. There were 2 replicates of 10 individuals for

each concentration. Mortality was assessed after 24 and 48 h. All tests were conducted at

20+1° C under 12:12 light:dark photoperiod, and organisms were not fed during the tests.

1.3.4 Reference toxicant tests

Reference toxicity tests were conducted with glochidia of L. fasciola, E.

capsaeformis and E. brevidens. Sodium chloride (NaCl) was used as the toxicant because

it is the suggested contaminant for reference bioassays with standard freshwater regulatory

11test organisms (US EPA 1993). A 0.5 serial dilution was used to create treatments, which

include a control, 0.5, 1.0, 2.0, 4.0, and 8.0 g NaCl/L diluent water; these are the same

concentrations for C. dubia reference tests. Certified reference grade NaCl was used as the

toxicant, and EPA100 water was used as the diluent, and control treatment. Viability of

glochidia was assessed after 24 and 48 h of exposure. Bioassays were conducted at 20+1°

C under a 12:12 light:dark photoperiod.

Results of monthly acute NaCl reference toxicant tests at the VPI&SU Aquatic

Toxicology Laboratory for NPDES permit tests with C. dubia, D. magna, and P. promelas

were compiled for comparative purposes. Tests were conducted between January 2001

and August 2003 according to standard protocol (US EPA 1993).

1.3.5 Water chemistry and Hg analysis

Temperature was monitored twice daily. Dissolved oxygen, conductivity, and pH

were measured for all in-water and out-water in the bioassays. Alkalinity and hardness

were measured for the control and highest concentration for in-water. An Accumet®

(Fisher Scientific, Pittsburgh, PA, USA) pH meter with an Accumet gel-filled combination

electrode (accuracy < + 0.05 pH at 25° C) was used to measure pH. Dissolved oxygen and

conductivity were measured with a 54A meter® and model 30 conductivity meter®,

respectively, from Yellow Springs (Yellow Springs, OH, USA). Total hardness and

alkalinity (as mg/L CaCO3) were measured in accordance with APHA et al. (1998) through

colorimetric titrations.

Samples of in- and out-water from several replicates were combined for each

treatment, and prepared for Inductively Coupled Plasma (ICP) spectrometry according to

US EPA (1991) standard methods. Trace metal-grade pure hydrochloric acid was used to

12reduce the sample pH to < 2. The prepared samples were refrigerated until analysis at the

VA Tech Soil Laboratory (Blacksburg, VA).

1.3.6 Data analysis

Toxicity test results were presented as LC50 values, and were calculated by

Spearman Karber analysis on computer software (Gulley 1993). All calculations were

based on nominal total Hg concentrations as treatments < 15 µg Hg/L were below detection

limits.

1.4 Results

1.4.1 Control survivorship

The combined mean glochidia viability in control treatments for all of the bioassays

was >89% for the species tested after 24 h (Fig. 1.1 [28]). Mean control survivorship

remained >80% after 48 h for all species, except L. fasciola, which declined to 78%.

Overall, viability substantial decreased with increased exposure time for all species except

V. iris.

1.4.2 Hg salt results

1.4.2.1 Mercuric chloride

Glochidia from the different species of freshwater mussels had similar tolerances to

MC, as 24-h and 48-h LC values for L. fasciola, E. capsaeformis, and E. brevidens ranged

between 25 – 54 and 27 - 40 µg Hg/L, respectively (Table 1.1 [23]). Although not evident

by the LC50 values, viability decreased with increased exposure time in nearly every

treatment. Survivorship remained high in the control (24 h =>89% and 48 h =>81%), but

was substantially reduced in treatments containing elevated concentrations of Hg. After 48

h, 100% mortality was observed in treatments >120 µg Hg/L.

13 Ceriodaphnia was far more sensitive to MC than D. magna, as the respective 48-h

LC50 values were 7 and 19 µg Hg/L (Table 1.2 [24]). Sensitivity increased with exposure

time in both tests, and the largest contrast in 24- and 48-h LC50 values (90 and 15 µg

Hg/L) was observed with D. magna. Survivorship in the control remained 100%, but was

substantially reduced in treatments with measurable concentration of Hg for both species.

1.4.2.2 Methyl mercuric chloride

The LC50 values for glochidia of E. capsaeformis and E. brevidens exposed to

MMC were substantially lower than those documented in MC tests. The LC50 values after

24 h ranged between 21 – 26 µg Hg/L for the two species (Table 1.3 [25]). However, 48-h

LC50 values could not be calculated because mortality was >50% in the lowest test

treatment, 8 µg Hg/L. Therefore, these values were reported conservatively as < 8 µg

Hg/L. Villosa iris glochidia were far more tolerant than the two other species. A 24-h

LC50 could not be calculated because only 38% of the individuals exposed to 120 µg Hg/L

died; however, the value was reported as > 120 µg Hg/L for comparative purposes. After

48 h, the LC50 for V. iris declined substantially to 43 µg Hg/L, but was still 5 times higher

compared to the values found for glochidia of the other species.

Ceriodaphnia was the most sensitive organism tested to MMC, as 100% mortality

occurred in treatments >8 µg Hg/L, despite 100% survivorship in the control (Table 1.4

[26]). The 48-h LC50 could not be calculated in either C. dubia test because of high

mortality in low concentrations. Subsequently, these values were reported conservatively

as < 4 and < 8 µg Hg/L. Daphnia were also quite sensitive to MMC, as 48-h LC50 values

for the two trials were 18 and 15 µg Hg/L. Pimephales promelas was extremely tolerant to

MMC exposure, as a 24-h LC50 values could not be calculated due to only 15% mortality

in the highest treatment; this value was expressed as > 120 µg Hg/L. A 48-h LC50 value

14was calculated for P. promelas that was considerably lower, 67 µg Hg/L, but remained

markedly higher than values for the other species.

1.4.3 Reference toxicant results

1.4.3.1 Glochidia

Glochidia of L. fasciola, E. capsaeformis, and E. brevidens had similar tolerances to

NaCl, as the upper and lower 95% confidence limits for the 24- and 48-h LC50 values

nearly overlapped (Table 1.5 [27]). After 48 h, control survivorship was extremely low in

the L. fasciola bioassay (68%) and therefore, the reported LC50 value of 2.25 g NaCl/L is

considered unreliable. Control survivorship was more stable in the bioassays with the other

species. As in the Hg tests, viability decreased with increased exposure times in most

treatments. The average 48-h LC50 for glochidia from all three species combined was 2.46

g NaCl/L (Fig 1.2 [29]).

1.4.3.2 Standard regulatory test organisms

The three different standard regulatory organisms, C. dubia, D. magna, and P.

promelas, had very distinct NaCl tolerances. The most sensitive species was C. dubia, as

the average 48-h LC50 was 2.33 g NaCl/L. Similar values for D. magna and P. promelas

were 4.96 and 9.84 g NaCl/L (Fig. 1.2 [29])

1.4.4. Water chemistry and Hg concentrations

Water chemistry parameters and Hg concentration analysis results for the different

test treatments are summarized in Table 1.6 [30]. Dissolved oxygen remained > 5.0 mg/L

in all bioassays. Other water parameters for in- and out-water did not differ substantially,

except for total Hg concentration, which was substantially lower in out-water. Treatments

15< 15 µg Hg/L were below detection limit (BDL). The in-water values for treatments > 30

µg Hg/L were very close to nominal concentrations.

1.5 Discussion

1.5.1 Mercury tests

Researchers have noted that glochidia may sporadically clasp, or completely seal

their valves when exposed to contaminants during laboratory toxicity tests. Effects induced

by toxicants may also be less apparent if glochidia remain open. However, researchers can

infer the viability of these individuals by exposing them to a noxious substance, such as

NaCl, that is known to elicit this avoidance behavior (Huebner and Pynnonen 1992,

Goudreau et al. 1993, Jacobson et al. 1997, and Keller and Ruessler 1997). Several

laboratory studies have reported that released glochidia have substantially lower viability in

treatments containing elevated concentrations of contaminants (Huebner and Pynnonen

1992, Goudrea et al. 1993, Jacobson et al. 1997, Keller and Ruessler 1997, Cherry et al.

2002). These observations have encouraged speculation that contaminants may be

attributing to the lack of recruitment in the water column by reducing the ability of

glochidia to successfully attach to host fish. A decrease in this ability would inevitably

lower reproductive potential of impacted individuals. In my study, very low concentrations

of Hg drastically affected the viability of glochidia. However, additional research is

needed to more accurately determine species tolerances because impairment was observed

in treatments with Hg concentrations below detection limit (BDL) in this study.

Regardless, this study provided substantial evidence that released glochidia from some

species of freshwater mussels are sensitive to Hg at measurable concentrations in the

environment.

16Interestingly, I also noted substantial inter-specific variability in Hg tolerances

among glochidia of different species, as individuals from L. fasciola, E. capsaeformis, and

E. brevidens were highly sensitive to acute exposure, but those from V. iris were not. Few

studies have conducted experiments with glochidia from numerous species of freshwater

mussels, and those report substantial variability in species tolerances. Cherry et al. (2002)

examined the effect of copper on glochidia from eight species of mussels, and reported

mean LC50 values that ranged from 37 to 137 µg Cu /L. Keller and Ruessler (1997)

conducted experiments on glochidia of six mussel species with the pesticide malathion, and

reported an even greater range, as the 48-h LC50 value for the most sensitive species,

Lampsilis siliquoidea was 7 mg/L, compared to 324 mg/L for the most tolerant species

tested, Utterbackia imbecillis.

There was also a distinct difference in the toxicity of the different Hg salt forms, as

glochidia from L. fasciola, E. capsaeformis, and E. brevidens were far more sensitive to

MMC than to MC. Although the same total Hg concentrations were tested for both salt

treatments, the 48-h LC50 values were approximately three times lower for glochidia

exposed to MMC. Several other studies have documented similar differences in the

toxicity of Hg salts with test organisms other than glochidia. Baby and Menon (1987)

observed that juvenile marine bivalves were more sensitive to Hg, if exposed to the organic

salt (CH3COO)2 Hg, than when exposed to the inorganic salt HgCl2. Similarly, Wobeser

(1975) reported that MMC is more toxic, and accumulates faster than MC in the tissues of

young age-classes of rainbow trout. Biesinger et al. (1982) cited that D. magna in a

chronic study excreted Hg slower when exposed to MMC than MC.

171.5.2 Glochidia tolerance compared to standard test organisms

Both C. dubia and D. magna were more sensitive to MC than glochidia of all

species tested, as 48-h LC50 values for standard organisms were 7 and 19, respectively,

compared to a range of 27-40 µg Hg/L for glochidia. The GMAV of 28 aquatic organisms

exposed to MC are documented in the 1984 WQC for Hg (US EPA 1985), and of them,

only 4 genera had mean LC50 values lower than the GMAV of 30 µg Hg/L for Epioblasma

in my study. The documented GMAVs for the more sensitive genera are 2.6 for Daphnia,

10 for Gammarus (amphipod), 20 for Chironomus (midge), and 20 µg/L for Faxonella

(crayfish). The GMAVs for other standard test organisms Salmo gairdneri (rainbow trout)

and P. promelas were 275 and 159 µg/L, respectively.

Both glochidia and standard regulatory organisms were more sensitive to MMC

than MC. Methyl-mercury is more toxic than other forms because it is more available to

biota and accumulates in aquatic food webs. Inorganic mercury is converted into methyl-

mercury through microbial respiration in aquatic systems, and we assume that this

conversion occurred in our test chambers. Though we did not measure MeHg

concentrations, MMC salt may have been more toxic than MC salt because a greater

portion of the total measured Hg in solution already existed in the organic form. It will be

important to have lower detection limits, and analyze both total Hg and MeHg

concentrations in future bioassays. Overall, this study suggests that aquatic organisms are

highly sensitive to Hg, regardless of the salt form being tested.

1.5.3 NaCl glochidia reference test

A dose-dependent response was evident in all of the glochidia reference tests, as

viability was substantially reduced in treatments with higher NaCl concentrations.

Furthermore, glochidia were quite sensitive to NaCl, as LC50 values were near those

18recorded for C. dubia, which is currently regarded as one of the more sensitive standard test

organisms. These observations support the use of NaCl as a reference toxicant for

glochidia, but additional studies are needed to verify these results, and improve the

precision of acceptable tolerance ranges for species.

Standard methods (US EPA 1991) require that reference toxicant tests be conducted

concurrently with WET testing for both acute and chronic C. dubia and P. promelas

bioassays. For results to be valid, endpoints of reference toxicant bioassays must be within

an acceptable range for a given species. Researchers are able to infer the relative health of

test organisms by comparing reference toxicity endpoints to established data bases based

on the premise that healthy individuals of a species will have similar tolerances to a

toxicant. Reference test endpoints that are below the acceptable species ranges suggest that

organisms used in the test may have inferior health, or that test conditions were not

acceptable.

Results are only acceptable for standard test organisms if a certain control

survivorship is maintained throughout tests; these thesholds are typically > 90 % for acute,

and > 80% for chronic bioassays (US EPA 1994, 1994). Additional requirements have also

been established for chronic C. dubia bioassays, and include a minimum control

survivorship of > 80%, > 60% of organisms in control treatments have three broods within

eight days, and surviving organisms in controls average >15 neonates. Chronic P.

promelas bioassay test results are only acceptable if control organisms survivorship is >

80%, and average growth is > 2.5 mg/individual. These thresholds have been established

by compiling the results of numerous trials, and provide researchers with specific

thresholds.

19 Several researchers have proposed similar validity endpoints for glochidia

bioassays, such as >90% after extraction, or >80% during the duration of the test (Keller

1993, Jacobson et al. 1997). Although these endpoints are useful and may potentially be

incorporated into protocols for glochidia bioassays, additional QA/QC measures are

essential for validating test results. I advocate conducting reference bioassays

concurrently with other glochidia toxicant tests as a means for inferring the relative health

of organisms used. Currently, additional research to determine the acceptable range in

NaCl tolerances of glochidia for a species of mussel is needed before reference test

endpoints can be used effectively as QA/QC measures. Since seasonal variation is

anticipated, it will also be important to conduct tests at different times throughout the year

to evlauted temporal effects.

1.6 Acknowledgment

The authors thank John Schmerfeld, Cindy Kane, and Jess Jones for their active

roles in the project. The research was supported by the U.S. Fish and Wildlife Service.

201.7 References

1) American Public Health Association (APHA), American Water Works Association (AWWA), and Water Environmental Federation (WEF). 1998. Standard Methods for the Examination of Water and Wastewater, 20th Edition. American Public Health Association, Washington, DC.

2) Augspurger T, AE Keller, MC Black, WG Cope, FJ Dwyer. 2003. Water quality guidance

for protection of freshwater mussels (Unionidae) from ammonia exposure. Environ Tox Chem 22: 2569-2575.

3) Baby KV and NR Menon. 1987. Salt forms of metals & their toxicity in the brown

mussel, Perna indica (Kuriakose & Nair). Indian J Mar Sci 16: 107-109. 4) Barkay T, M Gillman, RR Turner. 1997. Effects of dissolved organic carbon and salinity

on bioavailability of mercury. Applied Environ Micro: 4267-4271.

5) Beckvar N, S Salazar, M Salazar, K Finkelstein. 2000. An in situ assessment of mercury contamination in the Sudbury River, Massachusetts, using transplanted freshwater mussels (Elliptio complanata). Can J Fish Aquat Sci 57: 1103-1112.

6) Biesinger KE, LE Anderson, JG Eaton. 1982. Chronic effects of inorganic and organic

mercury on Daphnia magna: Toxicity, accumulation, and loss. Arch. Environ Contam Toxicol 11: 769-774.

7) Brenderman SA, Neves RJ. 1993. Life history of the endangered fine-rayed pigtoe, Fusconaia cuneolus (Bivalvia: Unionidae), in the Clinch River, Virginia. Am Malacol Bull 10:83-91.

8) Cherry DS, Van Hassel JH, Farris JL, Soucek DJ, Neves RJ. 2002. Site-specific derivation

of the acute copper criteria for the Clinch River, Virginia. Hum Ecol Risk Assess 8:591-601.

9) Diamond JM, DW Bressler, VB Serveiss. 2002. Assessing relationships between human land use and the decline of native mussels, fish, and macroinvertebrates in the Clinch and Powell River Watershed, USA. Environ Toxicol Chem 21: 1147-1155.

10) French KJ, DA Scruton, MR Anderson, DC Schneider. 1999. Influence of physical and

chemical characteristics on mercury in aquatic sediment. Water Air Soil Pollut 110: 347-362.

11) Goudrea SE, RJ Neves, RJ Sheehan. 1993. Effects of wastewater treatment plant effluents

on freshwater mollusks in the upper Clinch River, Virginia, USA. Hydrobiologia 252: 211-230.

12) Gulley DD. 1993. Toxstat Version 3.3. University of Wyoming, Department of Zoology

and Physiology, Laramie, WY.

2113) Henley WF, Neves RJ. 1999. Recovery status of freshwater mussels (Bivalvia:

Unionidae) in the North Fork Holston River, Virginia. Am Malacol Bull 15:65-73. 14) Huebner JD, KS Pynnonen. 1992. Viability of glochidia of two species of Anodonta

exposed to low pH and selected metals. Can J Zool. 70: 2348-2355.

15) Jacobson PJ, Neves RJ, Cherry DS, Farris JL. 1997. Sensitivity of glochidial stages of freshwater mussels (Bivalvia: Unionidae) to copper. Environ Toxicol Chem 16:2384-2392.

16) Keller, AE. 1993. Acute toxicity of several pesticides, organic compounds and wastewater effluent to the freshwater mussel, Anodonta imbecillis, Ceriodaphnia dubia and Pimephales promelas. Bull Environ Contam Toxicol 51:696-702.

17) Keller AE, Ruessler DS. 1997. The toxicity of malathion to unionid mussels: Relationship

to expected environmental concentrations. Environ Toxicol Chem 16:1028-1033.

18) Keller AE, Zam SG. 1991. The acute toxicity of selected metals to the freshwater mussel, Anodonta imbecillis. Environ Toxicol Chem 10:539-546.

19) Masnado RG, SW Geis, WC Sonzogni. 1995. Comparative acute toxicity of a synthetic

mine effluent to Ceriodaphnia dubia, larval fathead minnow and the freshwater mussel, Anodonta imbecillis. Environ Toxicol Chem 14: 1913-1920.

20) Mason RP, JM Laporte, S Andres. 2000. Factors controlling the bioaccumulation of mercury, methylmercury, arsenic, selenium, and cadmium by freshwater invertebrates and fish. Arch Environ Contam Toxicol 38: 283-297.

21) Mauro JBN, JRD Guimaraes, H. Hintelmann. 2002. Mercury methylation in macrophytes, periphyton, and water- comparative studies with stable and radio-mercury additions. Applied Bio Chem 374: 983-989.

22) Milam CD, Farris JF. 1998. Risk identification associated with iron-dominated mine discharges and their effect upon freshwater bivalves. Environ Toxicol Chem 17:1611-1619.

23) Moore DRJ, RS Teed, GM Richardson. 2003. Derivation of an ambient water quality criterion for mercury: Taking account of site-specific conditions. Environ Tox Chem 22: 3069-3080.

24) Mummert AK, RJ Neves, TJ Newcomb, DS Cherry. 2003. Sensitivity of juvenile freshwater mussels (Lampsilis fasciola, Villosa iris) to total and un-ionized ammonia. Environ Tox Chem 22: 2545-2553.

25) Naimo TJ. 1995. A review of the effects of heavy metals on freshwater mussels. Ecotoxicology 4:341-362.

26) Neves RJ, Widlak JC. 1987. Habitat ecology of juvenile freshwater mussels (Bivalvia: Unionidae) in a headwater stream in Virginia. Am Malacol Bull 5:1-7.

2227) Olson GF, DI Mount, VM Snarski, TW Thorslund. 1975. Mercury residues in fathead

minnows, Pimephales promelas Rafinesque, chronically exposed to methylmercury in water. Bull Environ Contam Tox. 14: 129-134.

28) US Environmental Protection Agency. 1985. Ambient water quality criteria for mercury- 1984. EPA 440/5-84-026. Washington, DC.

29) US Environmental Protection Agency. 1990. Proposed guide for conducting acute toxicity tests with the early life stages of freshwater mussels. EPA/68/02/4278. Washington, DC.

30) US Environmental Protection Agency. 1991. Methods for determining metals in environmental samples. EPA/600/4-91/010. Washington, DC.

31) US Environmental Protection Agency. 1993. Methods for measuring the acute toxicity of effluents and receiving waters to freshwater and marine organisms- 4th ed. EPA/600/4-90/027F, Washington, DC.

32) US Environmental Protection Agency. 1994. Short-term methods for estimating the chronic toxicity of effluents and receiving waters to freshwater organisms-3rd ed. EPA/600/4/91/002. Washington, DC.

33) Webber HM, TA Haines. Mercury effects on predator avoidance behavior of a forage fish, golden shiner (Notemigonus crysoleucas). Environ Tox Chem 22: 1556-1561.

34) Weinstein JE. 2001. Characterization of the acute toxicity of photoactivated fluoranthene to glochidia of the freshwater mussel, Utterbackia imbecillis. Environ Tox Chem 20: 412-419.

35) Weinstein JE. 2002. Photoperiod effects of the UV-induced toxicity of fluoranthene to freshwater mussel glochidia: Absence of repair during dark periods. Aquat Toxicol 59: 153-161.

36) Weinstein JE, Polk KD. 2001. Phototoxicity of anthracene and pyrene to glochidia of the freshwater mussel Utterbackia imbecillis. Environ Toxicol Chem 20:412-419

37) Wiener JG, PJ Shields. 2000. Mercury in the Sudbury River (Massachusetts USA): pollution history and synthesis of recent research. Can J Fish Aquat Sci 57: 1053-106120:2021-2028.

38) Williams JD, Warren ML, Cummings KS, Harris JL, Neves RJ. 1993. Conservation status of freshwater mussels of the United States and Canada. Fisheries 18:6-22.

39) Wobeser G. 1975. Acute toxicity of methyl mercury chloride and mercuric chloride for rainbow trout (Salmo gairdneri) fry and fingerlings. J Fish Res Board Can 32:2005-2013.

40) Yeager MM, RJ Neves, DS Cherry. 1999. Competitive interactions between early life stages of Villosa iris (Bivalvia: Unionidae) and adult Asian clams (Corbicula fluminea). Proceeding of the First Freshwater Mollusk Conservation Society Symposium.

Table 1.1. The 24- and 48-h LC50 values, respective 95% confidence intervals, and mean survivorship of glochidia of Lampsilis fasciola, Epioblasma capsaeformis, and Epioblasma brevidens exposed to different concentrations of mercuric chloride (HgCl2).

Species Concentration

(µg Hg/L) n 24-h

% Mortality24-h LC50 (95% CI) n

48-h % mortality

48-h LC50 (95% CI)

Control 200 6 200 19

5 200 4 200 17

10 200 4 200 15

L. fasciola 15 200 6 40 µg Hg/L 200 16 40 µg Hg/L

30 200 7 (40 – 50) 200 10 (30 - 40)

60 200 9 200 30

120 200 85 200 100

250 200 100 200 100

Control 200 3

8 200 4

L. fasciola 15 200 13 40 µg Hg/L n/a n/a n/a

30 200 60 (30 - 40)

60 200 100

120 200 100

Control 50 4 50 18

8 50 6 50 10

E. capsaeformis 15 50 16 25 µg Hg/L 50 36 27 µg Hg/L

30 50 64 (22 - 25) 50 68 (n/a)

60 50 100 50 100

120 50 100 50 100

Control 100 3 100 10

8 100 6 100 7

E. capsaeformis 15 100 8 54 µg Hg/L 100 6 36 µg Hg/L

30 100 14 (49 – 60) 100 28 (33 - 38)

60 100 50 100 95 120 100 100 100 100

Control 100 11 100 17

8 100 8 100 21

E. brevidens 15 100 12 47 µg Hg/L 100 16 27 µg Hg/L

30 100 17 (42 - 53) 100 53 (24 - 30)

60 100 62 100 100

120 100 100 100 100

23

Table 1.2. The 24- and 48-h LC50 values, respective 95% confidence intervals, and mean survivorship of Ceriodaphnia dubia and Daphnia magna exposed to different concentrations of mercuric chloride (HgCl2).

Species Concentration

(µg Hg/L) n 24-h

% Mortality24-h LC50 (95% CI) n

48-h % mortality

48-h LC50 (95% CI)

Control 20 0 20 0

4 20 5 20 15

C. dubia 8 20 30 11 µg Hg/L 20 60 7 mg Hg/L

15 20 60 (10 – 12) 20 85 (5 - 9)

30 20 100 20 100

60 20 100 20 100

Control 20 0 20 0

8 20 0 20 5

D. magna 15 20 5 90 µg Hg/L 20 40 19 µg Hg/L

30 20 5 (80 - 100) 20 80 (17 - 22)

60 20 15 20 100

120 20 80 20 100

24

Table 1.3. The 24- and 48-h LC50 values, respective 95% confidence intervals, and mean survivorship of glochidia of Epioblasma capsaeformis, Epioblasma brevidens, and Villosa iris exposed to different concentrations of methyl-mercuric chloride (CH3HgCl2).

Species Concentration

(µg Hg/L) n 24-h %

Mortality 24-h LC50 (95% CI) n

48-h % mortality

48-h LC50 (95% CI)

Control 50 4 50 18

8 50 10 50 70

E. capsaeformis 15 50 36 21 µg Hg/L 50 80 8 µg Hg/L

(individual A) 30 50 68 (17- 24) 50 100 (4 - 9)

60 50 100 50 100

120 50 100 50 100

Control 100 3 100 10

8 100 4 100 49

E. capsaeformis 15 100 13 26 µg Hg/L 100 100 < 8 µg Hg/L

(individual B) 30 100 60 (23 - 28) 100 100 (n/a)

60 100 100 100 100

120 100 100 100 100

Control 100 11 100 17

8 100 10 100 56

E. brevidens 15 100 26 25 µg Hg/L 100 100 < 8 µg Hg/L

30 100 51 (22 - 28) 100 100 (n/a)

60 100 100 100 100

120 100 100 100 100

Control 326 6 305 5

8 246 4 316 5

V. iris 15 257 6 > 120 µg Hg/L 309 8 43 µg Hg/L

30 316 6 325 15 (41 - 45)

60 276 8 314 90 120 255 38 336 100

25

Table 1.4. The 24- and 48-h LC50 values, respective 95% confidence intervals, and mean survivorship of Ceriodaphnia dubia, Daphnia magna, and Pimephales promelas exposed to different concentrations of methyl-mercuric chloride (CH3HgCl2).

Species Concentration

(µg Hg/L) n 24-h %

Mortality 24-h LC50 (95% CI) n

48-h % mortality

48-h LC50 (95% CI)

Control 20 0 20 5.0

8 20 10 20 100

C. dubia 15 20 15 30 µg Hg/L 20 100 < 8 µg Hg/L

30 20 60 (20 - 30) 20 100 (n/a)

60 20 90 20 100

120 20 100 20 100

Control 20 0 20 0

4 20 5 20 85

C. dubia 8 20 15 25 µg Hg/L 20 100 < 4 µg Hg/L

15 20 15 (20 - 30) 20 100 (n/a)

30 20 30 20 100

60 20 100 20 100

Control 20 0 20 0

8 20 0 20 5.0

D. magna 15 20 0 20 µg Hg/L 20 15 18 µg Hg/L

30 20 95 (20 - 22) 20 100 (15 - 21)

60 20 100 20 100

120 20 100 20 100

Control 20 0 20 0

4 20 0 20 0

D. magna 8 20 0 > 60 µg Hg/L 20 5 15 µg Hg/L

15 20 5 20 45 (11 – 19)

30 20 15 20 100

60 20 35 20 100

Control 20 0 20 0

0.008 20 0 20 0

P. promelas 0.015 20 0 120 µg Hg/L 20 0 67 µg Hg/L

0.03 20 0 (n/a) 20 0 (57 - 77)

0.06 20 0 20 35

0.12 20 15 20 100

26

Table 1.5. The 24- and 48-h LC50 values, respective 95% confidence intervals, and percent mortality of glochidia of Lampsilis fasciola, Epioblasma capsaeformis, and Epioblasma brevidens exposed to the reference toxicant NaCl.

Species

Concentration (g NaCl/L)

Temp (°C)

n

24-h % Mortality

24-LC50 (95% CI)

n

48-h % Mortality

48-h LC50 (95% CI)

Control 22 200 3.5 200 32 0.5 22 200 3.0 200 18

L. fasciola 1.0 22 200 2.0 3.08 g NaCl/L 200 23 2.25 g NaCl/L 2.0 22 200 17.5 (2.91-3.26) 200 49.5 (2.14-2.35) 4.0 22 200 73.5 200 100 8.0 22 200 100 200 100 Control 20 50 6.0 50 8.0 0.5 20 50 4.0 50 6.0

E. capsaeformis 1.0 20 50 4.0 2.71 g NaCl/L 50 8.0 2.45 g NaCl/L 2.0 20 50 8.0 (2.54-2.88) 50 22 (2.21-2.70) 4.0 20 50 100 50 100 8.0 20 50 100 50 100 Control 20 50 8.0 100 14 0.5 20 50 6.0 100 12

E. brevidens 1.0 20 50 6.0 2.68 g NaCl/L 100 12 2.67 g NaCl/L 2.0 20 50 10 (2.50-2.88) 100 10 (2.54-2.79) 4.0 20 50 100 100 100 8.0 20 50 100 100 100

27

0

10

20

30

40

50

60

70

80

90

100

Time (h)

24 48

Mean survivorship (%)

Figure 1.1. Mean viability of mussel glochidia from VillEpioblasma brevidens, and Lampsilis fasciola in control

Villosa iris Epioblasma capsaeformis Epioblasma brevidens Lampsilis fasciola

osa iris, Epioblasma capsaeformis, treatments after 24 and 48 h.

28

Jan

01Fe

b 01

Mar

01

Apr 0

1M

ay 0

1Ju

ne 0

1Ju

ly 0

1Au

g 01

Sept

01

Oct

01

Nov

01

Dec

01

Jan

02Fe

bv 0

2M

ar 0

2Ap

r 02

May

02

June

02

July

02

Aug

02Se

pt 0

2O

ct 0

2N

ov 0

2 D

ec 0

2Ja

n 03

Febv

03

Mar

03

Apr 0

3M

ay 0

3Ju

ne 0

3Ju

ly 0

3Au

g 03

LC50

val

ues

(g N

aCl/L

)

0

2

4

6

8

10

12

Ceriodaphnia dubia Daphnia magnaPimephales promelas

Figure 1.2. The48-h LC50 values of reference toxicant tests conducted with standard regulatory test organisms Ceriodaphnia dubia, Daphnia magna, and Pimphales promelas. Bioassays conducted according to US EPA Protocol (1999)

29

Table 1.6. Mean conductivity, pH, Alkalinity, Hardness, and in- and out-water mercury concentrations for the acute mercury and reference toxicant tests.

Test Treatment

Conductivity

(µmhos)

pH

(su)

Alkalinity

(mg/L as CaCO3)

Hardness

(mg/L as CaCO3)

In Hg

[µg/L]

Out Hg

[µg/L]

All Control 298 + 9 7.78 + 0.13 62.7 + 4.4 83.5 + 4.6 BDL BDL

4 297 + 8 7.81 + 0.11 n/a n/a BDL BDL

8 300 + 3 7.80 + 0.09 n/a n/a BDL BDL

MC 15 294 + 14 7.77 + 0.14 n/a n/a BDL BDL

30 296 + 8 7.83 + 0.06 n/a n/a 31.4 + 8.0 26.6 + 9.0

60 301 + 5 7.76 + 0.18 n/a n/a 63.2 + 7.1 52.7 + 18

120 297 + 11 7.81 + 0. 08 63.1 + 5.7 86.2 + 7.8 117.7 + 21.6 98.4 + 27

4 297 + 12 7.78 + 0.08 n/a n/a BDL BDL

8 302 + 18 7.82 + 0.14 n/a n/a BDL BDL

MMC 15 293 + 7 7.81 + 0.12 n/a n/a BDL BDL

30 298 + 5 7.79 + 0.09 n/a n/a 32.9 + 7.3 22.5 + 14

60 299 + 11 7.81+ 0.14 n/a n/a 62.0 + 18.4 46.7 + 22

120 294 + 9 7.83 + 0. 12 61.9 + 7.2 84.8 + 5.6 133.6 + 38.8 86.6 + 41

0.5 1218 + 104 7.84 + 0.06 n/a n/a n/a n/a

1 2154 + 131 7.82 + 0.04 n/a n/a n/a n/a

NaCl 2 3884 + 248 7.84 + 0.11 n/a n/a n/a n/a

4 7160 + 177 7.83 + 0.12 n/a n/a n/a n/a

8 14570 + 342 7.81 + 0.13 62.6 + 3.9 82.9 + 5.8 n/a n/a

30

Chapter 2: The use of sodium chloride as a reference toxicant for evaluating the

quality of freshwater mussel glochidia.

31

2.1 Abstract

Glochidia have been proposed as a surrogate life stage for juvenile mussels in bioassays

because of their greater availability and demonstrably lower tolerances to some

contaminants. However, the validity of glochidia test results are under scrutiny by

regulatory agencies for lack of quality assurance / quality control (QA/QC) measures, and

concerns about assessing the health of test organisms. Sodium chloride was used as a

reference toxicant to assess the health of freshwater mussel glochidia. Glochidia from

four mussel species were tested, and LC50 values were significantly different among

species; respective means were 2.5 g NaCl/L for Alasmidonta heterodon, 1.7 g NaCl/L

for Epioblasma capsaeformis, 2.1 g NaCl/L for Lampsilis fasciola, and 0.98 g NaCl/L for

Villosa iris after 24 or 48 h of exposure. Mean control survivorship remained > 90%

after 24 h, and > 80% after 48 h for all species. More importantly, I observed a high

degree of variability among LC50s for glochidia collected from different gravid adults of

the same species, despite minimal differences in control survivorship. The study also

provides tentative methods for conducting glochidia reference toxicant tests that may

serve as QA/QC measures in future tests.

Keywords: glochidia freshwater mussel sodium chloride quality viability

32

2.2 Introduction

The imperiled status of many freshwater mussel species has intensified interest in

identifying potential factors leading to their demise. For several decades, scientists have

recognized the value of using adult mussels as bio-indicators because of their sedentary

nature, close association with benthic and pelagic zones, and ability to tolerate the

accumulation of high levels of heavy metals in their tissue (Salanki and Balogh 1989,

(Metcalfe-Smith 1994, Warren et al. 1995, Malley et al. 1999, Beckvar et al. 2000).

However, not until recent advances in culturing techniques were sufficient numbers of

juvenile mussels available to conduct toxicity tests. It is well documented that early life

stages of most aquatic organisms are more sensitive to toxicants than adults, and this

trend is also apparent in studies focused on juvenile and adult mussels (Naimo 1995,

Jacobson et al. 1997, Keller Ruessler 1997). The discrepancy in sensitivities between the

two life stages may be due to differences in physiology (i.e., thinners shells that are more

permeable to toxicants), metabolism (i.e., faster accumulation and slower excretion rates

of toxicants relative to body mass), or behavior (i.e., adults rely primarily on filter-

feeding whereas juveniles rely on both filter- and pedal-feeding) (Yeager et al. 1994,

Naimo 1995).

Observations in rivers also indicate that juveniles have greater sensitivities than

adults, as recent mussel surveys have shown disproportionately low percentages of young

mussels in populations and lack of recruitment at sites of adequate habitat (Yeager et al.

1994, Henley and Neves 1999). This has led researchers to question whether current

United States Environmental Protection Agency (EPA) Water Quality Criteria (WQC)

are suitable for protecting the early life stages of mussels (Augspurger et al. 2003).

Additional studies are needed, but it has proven difficult and costly to obtain sufficient

33

numbers of juveniles for conducting tests, for lack of mussel culturing facilities in the

United States (Newton et al. 2003). These difficulties are exacerbated when threatened or

endangered species are involved. Therefore, researchers are considering glochidia, the

parasitic larval stage, as a surrogate for the juvenile life stage in toxicity tests because

greater numbers are more readily available (Valenti et al. 2003)

While in the water column, glochidia may be exposed to toxicants that impair

their ability to attach to host fish (Keller and Ruessler 1997), thus reducing reproductive

potential. Laboratory tests have shown that exposure to stressors may provoke a positive

response by glochidia, which includes sporadic snapping or complete closure of their

valves (Goudreau et al. 1993, Jacobson et al. 1997); however, this behavior is

undocumented in rivers. Researchers speculate that closed glochidia are unable to attach

to host fish, and therefore are functionally unavailable (Goudreau et al. 1993, Jacobson et

al. 1993, Weinstein and Polk 2001). Closed glochidia may not re-open, and the response

is likely influenced by the toxicant, concentration, and exposure time. The valve closure

response can be easily quantified, and has proven to be an effective means for assessing

glochidia viability in toxicological studies. Several researchers have used concentrated

NaCl solutions to elicit this behavior (Goudreau et al. 1993, Jacobson et al. 1997, Keller

and Ruessler 1997). Open glochidia that respond to NaCl by closing their valves are

considered viable, whereas those with closed valves prior to, or whose valves remain

gaped after the salt solution exposure, are classified as functionally dead.

Few studies have compared the relative sensitivities of glochidia and juveniles of

the same species, although Keller and Ruessler (1997) compared acute toxicity of the

insecticide malathion for both life stages of Utterbackia imbecillis, Villosa lienosa, and

V. villosa. The glochidia of the three species had lower 48-h LC50 values than newly

34

transformed juveniles, which is indicative of greater sensitivity. Valenti et al. (2003)

tested the sensitivities of V. iris glochidia and 2-month old juveniles to mercuric chloride

(HgCl2) and found 48-h LC50 values of 52 µg Hg/L and 135 µg Hg/L, respectively.

Furthermore, 24- and 48- h LC50 values in similar tests with 2-mo old Epioblasma

capsaeformis juveniles were 0.16 and 0.14 mg Hg/L, respectively, (DS Cherry,

unpublished data), which are substantially higher than those reported for glochidia (0.04

and 0.03 mg Hg/L, respectively) (Valenti et al. 2003). Augspurger et al. (2003) reported

V. iris and U. imbecillis glochidia to be equally or more sensitive to total ammonia in

acute laboratory studies. The greater sensitivities of glochidia noted in previous acute

studies further support their use as surrogate test organisms for juvenile mussels in short-

term exposure toxicity tests.

Cherry et al. (2002) compared the acute sensitivities of glochidia and standard test

organisms in a study to develop an acute site-specific Criterion Maximum Concentration

(CMC) for copper. Acute bioassays were conducted with 17 species to assess the

potential impact of a coal-burning power plant’s effluent on mussel populations in the

Clinch River, VA. They reported that glochidia from three mussel species were more

sensitive than Ceriodaphnia dubia and Pimephales promelas. The Genus Mean Acute

Values (GMAV) for glochidia of the four most sensitive mussel species ranged from 37-

60 mg/L, compared to 88 and 310 mg/L for C. dubia and P. pimephales, respectively. In

a study examining the toxicity of different mercury salt forms, Valenti et al. (2003)

reported sensitivities of glochidia of Epioblasma capsaeformis and E. brevidens to

methyl-mercuric chloride comparable to that of C. dubia, and greater than Daphnia

magna and P. promelas after 48-h exposures. Milam and Farris (1998) examined the

toxicity of partially treated mine water and reported glochidia of Leptodea fragilis to be

35

more sensitive than P. promelas, but less sensitive than D. magna and C. dubia.

However, their study contrasted 24-h acute glochidia LC50s with 48-h acute LC50s for

D. magna and 7-day fecundity EC50s for C. dubia. Additional studies have suggested

that glochidia are extremely sensitive when compared to other aquatic organisms

(Goudreau et al. 1993, Weinstein 2001, Weinstein and Polk 2001).

The suitability of using glochidia as test organisms for assessing environmental

impairment is currently under evaluation, and regulatory agencies are hesitant to accept

the results of previous studies for a lack of standardized test protocols and Quality

Assurance/ Quality Control (QA/QC) measures. Many of these concerns were expressed

after guidelines for conducting laboratory tests with early life stages of freshwater

mussels were proposed (US EPA 1990). The primary criticism of using glochidia as test

organisms was difficulty in accurately verifying organism health at test initiation. Test

validity endpoints previously suggested include criteria such as viability >90%

immediately after extraction, or control survivorship >80% over the test duration

(Jacobson et al. 1997, Keller and Ruessler 1997). Although these endpoints are useful

and may eventually be incorporated as principle components of acceptable glochidia

bioassay criteria, there is still concern for the accuracy and precision of these approaches.

Current U.S. EPA protocol for standard test organisms require not only a

minimum control survivorship (typically >90%), but also the completion of reference

toxicant tests to ensure adequate organism health. Reference toxicity tests are required 1)

monthly at test organism culturing facilities, and 2) concurrently with permit tests

conducted with organisms collected from the field or obtained from commercial suppliers

(US EPA 1993, 1994). Reference toxicant tests allow researchers to better infer the

relative health of test organisms, ensure suitability of test conditions, and verify that test

36

procedures were appropriate by contrasting reference test endpoints with those of an

established database. Overall, the current reference toxicant system for standard test

organisms aids in validating test results, and helps maintain uniformity among test

facilities.

2.2.1 Study Objectives

The study goal was to propose a potential reference toxicant test protocol for

glochidia supported by tests results, and to begin to develop tentative species data bases.

I hope to encourage future reference tests trials with glochida so that a quality control

assurance criterion can be established.

2.3 Materials and Methods

2.3.1 Test Organisms

Gravid adults of Villosa iris (rainbow mussel), Epioblasma capsaeformis (oyster

mussel), and Lampsilis fasciola (wavyrayed lampmussel) from the Clinch River, VA, and

Alasmidonta heterodon (dwarf wedgemussel) from the Ashuilot River, NH were

collected during the 2003 summer field season. The oyster mussel and dwarf

wedgemussel are federally endangered species. Adult mussels were held in a

recirculating trough at 21 + 1° C to acclimate them to laboratory conditions before the

glochidia were extracted. Glochidia were removed by gently prying open a gravid

female, puncturing the marsupium with a water-filled sterile syringe, and then injecting

the water to flush the glochidia.

2.3.2 NaCl Test Design

The glochidia reference test design was based on established US EPA protocols (1993)

for other freshwater organisms (Table 2.1 [54]). In the initial reference test with V. iris,

test concentrations were a 0, 0.5, 1.0, 2.0, 4.0, and 8.0 g NaCl/L. However, due to 100%

37

mortality at the two highest concentrations, the 8.0 g NaCl/L concentration was

eliminated in all subsequent tests and replaced with a 0.25 g NaCl/L concentration.

Moderately hard, synthetic water (US EPA 1993, 1994) served as diluent and control.

The reference toxicant was Certified A.C.S. grade NaCl salt (Fisher Scientific S271-3).

A 0.5 serial dilution was used to prepare the test concentrations. Bioassays were

conducted for 48 h, and viability was assessed after each 24-h interval. However, A.

heterodon bioassays were conducted for only 24 h because it was essential that most

glochidia of this endangered species be used to infest host fish for production of

juveniles. Similarly, one female E. capsaeformis partially released glochidia while in

captivity, leaving only sufficient glochidia for a 48 h test.

Bioassays were conducted at 20+1° C under a 16:8 light:dark photoperiod, with

eight replicates for each concentration. Four replicates were randomly sampled after 24 h

and glochidia viability was assessed, and the remaining four replicates were checked after

48 h. The test chambers consisted of 50-ml glass beakers filled with ~ 40 ml of test

solution. Glochidia (25-75 total) were loaded into each test chamber with a fine tip glass

pipette. Precautions were taken to evenly distribute glochidia in test chambers, to

minimize potential interactions among test organisms in the same replicate. To assess

viability, a sample of glochidia was drawn out of the test chamber with a fine-tip glass

pipette and transferred to a glass petri dish for observation with a dissecting microscope.

After the numbers of open and closed glochidia were recorded, they were then exposed to

a concentrated NaCl solution. Glochidia were recorded as functionally dead if they were

closed before or did not show a response after addition of the salt solution. A response

was defined as either the complete, partial, or sporadic closure of the valves.

38

2.3.3 Water chemistry

Temperature, dissolved oxygen, conductivity, and pH were measured at the

initiation and completion of each bioassay. Alkalinity and hardness were measured for

the control and highest concentration. An Accumet® (Fisher Scientific, Pittsburgh, PA,

USA) pH meter with an Accumet gel-filled combination electrode (accuracy < + 0.05 pH

at 25° C) was used to measure pH. Dissolved oxygen was measured with a Yellow

Springs model 54A meter® (Yellow Springs, Yellow Springs, OH, USA), and a Yellow

Springs model 30 conductivity meter® (Yellow Springs, Yellow Springs, OH, USA) was

used to measure specific conductivity (accuracy + 0.5%). Total hardness and alkalinity

(as mg/L CaCO3) were measured in accordance with APHA et al. (1998) through

colorimetric titrations. Incubator temperature was monitored twice daily.

2.3.4 Data analysis

Data from the two separate trials of testing with V. iris and E. capsaeformis were

pooled for statistical analysis, as there were no significant differences in control

survivorship or LC50 values between testing periods for either species (p<0.05).

Spearman-Karber LC50 values were calculated using the Toxstat® program. One-way

ANOVAs and Student’s t test pair-wise comparisons were calculated with the computer

program Jump In 4.0® (α = 0.05 level). Statistics were not computed for concentrations

> 4.0 g NaCl/L because mortality was 100% for all species.

39

2.4 Results

2.4.1 Intraspecific variation

2.4.1.1 Villosa iris

The 24-h control survivorship means for V. iris glochidia from the various adult

mussels ranged from 89.2 - 98.0% and were not significantly different (p=0.12) (Table

2.2 [55]). Glochidia from mussel # 8 had the lowest control survivorship (89.2%), which

was slightly lower than that of the other mussels (93.1-98.0%). However, the ANOVA

for all other concentrations < 4.0 g NaCl/L had significant variation in survivorship

(p<0.001).

The Student’s t pair-wise comparisons at the 0.5, 1.0, 2.0 g NaCl/L concentrations

had significant variation in mean survivorship among glochidia of different individuals

after 24 h of exposure. At the 0.5 g NaCl/L concentration, mean survivorship of

glochidia from mussel # 8 was approximately three-times lower than that of the other

seven individuals tested; 28.5% compared to 72.6 – 90.5%. Glochidia collected from

mussel #8 also had the lowest survivorship (20.5%) at the 1.0 g NaCl/L concentration.

Survival also was greatly reduced for glochidia from mussel #7 at this concentration, with

mean survivorship declining from 74.5% at 0.5 g NaCl/L to 41.0% at 1.0 g NaCl/L.

There was also significant variation in survivorship of glochidia from different

individuals at 2.0 g NaCl/L; however, mean viability was drastically reduced for all

glochidia (1.6 – 16.5%).

The 24-h LC50s also varied markedly for glochidia collected from different V. iris

females, as values ranged from 0.31-1.51 g NaCl/L. The LC50 of 0.31 g NaCl/L for

mussel #8 was 3-5 times lower than for the other seven individuals. Six of the eight

40

individuals had LC50s >1.0 g NaCl/L, and of these, five had confidence intervals that

nearly overlapped. The LC50 of 1.51 g NaCl/L for glochidia from mussel #1 did not

overlap, and was substantially higher than the values for glochidia from the other

females.

There was a highly significant difference in 48-h control survivorship among

glochidia from different individuals of V. iris (p=0.0016). The lowest mean survivorship

of 82.9% occurred with glochidia obtained from mussel #8, and was significantly lower

than the control survivorship means (89.6 – 97.0%) for the glochidia from the other

mussels. Overall, there was a significant difference between 24- and 48-h control

survivorship for glochidia of V. iris (p=0.045), as the means were 95.2 + 4.9% and 92.5 +

5.7%, respectively.

The 0.5, 1.0, and 2.0 g NaCl/L concentrations also had significant variations in

survivorship among glochidia collected from the various V. iris females after 48 h of

exposure. Similar trends to the 24-h data were apparent. Again, individual #8 had the

lowest survivorship at the 0.5 and 1.0 g NaCl concentrations, 29.0 and 6.0%,

respectively. Glochidia from individual #7 also had significantly impaired survivorship,

when compared to the other individuals at these concentrations (50.0 and 37.5%,

respectively), despite 95.0% control survivorship.

The 48-h LC50 values for glochidia from the eight female V. iris varied

substantially. An LC50 value could not be calculated for mussel #8 because mortality

was >50% at the lowest test concentration, so the value was expressed conservatively as

<0.25 g NaCl/L. Individual #7 had an LC50 of 0.57 g NaCl/L, which was 2-4 fold lower

than those recorded for the other six individuals. Though glochidia from individuals #5

and #6 had LC50 values < 1.0 g NaCl (0.99 and 0.97 g NaCl/L, respectively), the upper

41

95% confidence intervals exceeded this limit. The LC50s for glochidia from the

remaining four V. iris ranged from 1.15-1.60 g NaCl/L, and mussel #1 again had the

highest value. The 24- and 48-h LC50s were not significantly different for glochidia of

V. iris (p=0.66), as mean values were 1.07 + 0.36 and 0.98 + 0.41 g NaCl/L, respectively.

2.4.1.2 Epioblasma capsaeformis

The 24-h control mean survivorships for E. capsaeformis glochidia from the

various adults were not significant different (p=0.11), and ranged between 87.3 – 93.7%

(Table 2.3 [56]). Survivorship also was not significant different at the 0.25 and 0.5 g

NaCl/L concentrations (p= 0.063 and p=0.21, respectively); however, there were highly

significant differences at the 1.0 and 2.0 g NaCl/L concentrations (p= <0.0001 and p=

0.0003).

At the 1.0 g NaCl/L concentration, the mean survivorship for glochidia from

mussel #5 (46.3%) was approximately half the value recorded for glochidia from the

other individuals, which ranged from 86.9 – 92.2%. A similar trend was also apparent at

the 2.0 g NaCl/L concentration for glochidia from mussel #5, as mean survivorship was

only 11.5%; approximately 3-5 times lower than the values (35.9 – 50.9%) recorded for

glochidia from the other four mussels.

There was also variation in the 24-h LC50 values for glochidia of E.

capsaeformis. Glochidia from mussel #5 had a substantially lower LC50, 1.10 g NaCl/L,

compared to the other four mussels, 1.81 – 1.98 g NaCl/L. Notably, glochidia from

adults that maintained mean control survivorship >90% also had markedly higher LC50

values.

42

There were highly significant differences among control survivorship values for

glochidia of E. capsaeformis after 48 h of exposure to NaCl (p= 0.0069). Control

survivorship for glochidia from mussels # 3 (80.4%) and #4 (76.9%) were significantly

lower than the values recorded for mussels # 1 (91.5%) and #2 (89.1%). No comparison

could be made with mussel #5, which had the lowest control survivorship after 24 h,

because there were inadequate glochidia to run a 48-h test. Overall, control survivorship

was significantly reduced after 48 h compared to 24 h (p=0.0023), and the 24- and 48-h

means were 90.8 + 3.4 and 84.4 + 7.8%, respectively.

Statistically significant variations in survivorship of glochidia from different E.

capsaeformis females were noted at the 0.5, 1.0, and 2.0 g NaCl/L concentrations after 48

h. Mean survivorship for glochidia from mussel #4 was only 66.5%, compared to 86.6 -

88.9% for glochidia from the other three mussels. A similar trend was apparent at the 1.0

g NaCl/L concentration. The most striking variation in sensitivity among E.

capsaeformis glochidia was noted at the 2.0 g NaCl/L concentration as the survivorship

for glochidia from mussel #4 was significantly lower (4.2%), compared to the other four

mussels (24.8 – 74.4%).

The 48-h LC50 values for glochidia of E. capsaeformis also varied markedly.

The lowest LC50 recorded was for glochidia from mussel #4 (1.38 g NaCl/L), but the

values were higher for glochidia from the other mussels (1.65-1.90 g NaCl/L). Glochidia

from mussels #1-3 had LC50s with overlapping confidence limits and control

survivorship >80%. Overall, the 24- and 48-h LC50 values were not significantly

different for glochidia of E. capsaeformis, as means were 1.73 + 0.36 and 1.70 + 0.24 g

NaCl/L, respectively.

43

2.4.1.3 Alasmidonta heterodon

There was no significant difference in control survivorship of glochidia between

the two A. heterodon females (p=0.996), as means were 95.5 and 95.4%, respectively

(Table 2.4 [57]). Furthermore, survivorship for glochidia from the different adults did

not vary significantly at any concentration. The 24-h LC50s were 2.62 and 2.41 g

NaCl/L, and their 95% confidence limits overlapped.

2.4.1.4 Lampsilis fasciola

A significant difference was evident among 24-h glochidia control survivorship

from the three L. fasciola females (p=0.02) (Table 2.5 [58]). Glochidia from mussel #3

had a significantly lower mean control survivorship of 84.9% compared to 94.2 and

94.0% for the other two mussels. Significant variation in 24-h survivorship also occurred

among the glochidia at all concentrations < 2.0 g NaCl/L. Glochidia from mussel #3

consistently had the lowest survivorship at most concentrations, although the minimum

mean viability remained >80%. Conversely, mussel #3 had the highest mean

survivorship of 64.8% at the 2.0 g NaCl/L concentration; however, it was not

significantly different from the means (57.8 and 62.3%) for the glochidia from the other

mussels (p=0.86).

The 24-h LC50s ranged from 2.09 – 2.41 g NaCl/L. Glochidia from mussel #3

had the highest LC50 despite having the lowest control survivorship, which was atypical

of the pattern observed in our other experiments. However, this discrepancy was minor,

as 95% confidence limits nearly overlapped for glochidia from all three individuals.

After 48 h of exposure, mean control survivorship for glochidia from mussel #3

was 80.0%, and was significantly lower than values from the other mussels (87.5 and

44

91.3%). Lampsilis fasciola was the only species tested in which glochidia did not have a

significant difference in control survivorship between 24 and 48 h (p=0.06); however, a

larger sample size is needed to confirm this observation. The respective means were 91.0

+ 6.0 and 86.2 + 6.0%.

There were significant differences among the survivorships of glochidia from

different adult mussels at concentrations < 2.0 g NaCl/L, though no decisive trend was

apparent. The 48-h LC50s did not vary greatly for glochidia from different adults, and

values ranged from 2.00 – 2.30 g NaCl/L. The 24- and 48-h LC50s were not significantly

different (p=0.40), and means were 2.24 + 0.16 and 2.12 + 0.16 g NaCl/L, respectively.

2.4.2 Interspecific variation

2.4.2.1 Control Survivorship

The 24-h control survivorships for glochidia from the various species were

significantly different (p=0.0041). Student’s t pair-wise comparison revealed that V. iris

and A. heterodon had significantly higher control survivorships than E. capsaeformis and

L. fasciola glochidia (p=0.05). However, glochidia from all species averaged >90%

control survivorship after 24 h of exposure (Fig. 2.1 [59]). The results of the one-way

ANOVA contrasting 48-h control survivorship among species showed a highly

significant difference (p= 0.0002). Villosa iris was the only species that sustained a mean

control survivorship >90%. No 48-h data were available for A. heterodon.

2.3.2.2 LC50 values

The 24-h LC50s of glochidia from the four mussel species tested were

significantly different (p<0.0001) (Fig 2.1 [59]). The most NaCl-tolerant species was A.

heterodon (2.52 + 0.15 g NaCl/L), while the most sensitive species was V. iris (1.07 +

45

0.36 g NaCl/L). The average LC50 values for L. fasciola and E. capsaeformis were 2.24

+ 0.16 and 1.73 + 0.36 g NaCl/L, respectively. Forty-eight hour LC50s also varied

significantly among species (p=0.008), as glochidia of V. iris again had a significantly

lower value of 0.98 + 0.41 compared to 2.12 + 0.16 and 1.70 + 0.24 g NaCl/L for L.

fasciola and E. capsaeformis, respectively (p<0.05).

2.4.2.3 Water Chemistry

Temperature was 20+1° C, dissolved oxygen was >6.5 mg/L, and pH ranged from

7.84 – 8.19 in all treatments. Alkalinity and hardness ranged from 62-68 mg CaC03 and

82-86 mg CaC03, respectively. Conductivity (µS/cm) increased with concentration:

control = 264-357, 0.25 = 724-774, 0.5 = 1100- 1290, 1.0 = 2047-2340, 2.0 = 3740-

4086, 4.0 = 7050 – 7430, and 8.0 g NaCl/ L = 13910-15240.

2.5 Discussion

2.5.1 Intra-specific variation

The NaCl tolerances and control survivorships of glochidia collected from

different gravid females varied substantially within some species. Intra-specific

variability was most pronounced for glochidia of V. iris, as 48-h LC50 values ranged

from <0.25 – 1.60 g NaCl/L, and control survivorship varied from 82.9 – 97.0%. The

greater variability reported for V. iris may be attributed to the larger sample size (n=8) of

experiments. More tests were run with V. iris because they are a relatively common

mussel, making glochidia easier to obtain and expandable. Intra-specific variability will

likely become evident for other species if larger sample sizes are used in future tests.

The substantial intra-specific variability in survival observed in our study

confirms previous toxicological efforts that report a broad range in LC50 values for

46

glochidia from females of the same species. In experiments assessing the toxicity of

copper, Cherry et al. (2002) reported large ranges in LC50 values (µg Cu/L) for glochidia

of L. fasciola (26 – 51), V. iris (36 – 98), Medionidus conradicus (16 – 81),

Ptychobranchus fasciolaris (17 – 212), and Actinonaias ligamentina (30 –172).

Similarly, Jacobson et al. (1997) reported 24-h LC50 values (µg Cu/L) of 26 and 46 for

L. fasciola and 36, 37, 39, 46, 46, 65, 73, 75, and 80 for V. iris.

There are several factors that may affect the tolerances of glochidia and result in

intra-specific variation; of these, many are associated with obtaining gravid females from

different field sites. Habitat disturbances such as pollution, sedimentation, reduction in

food quality/availability, and flood events may affect the health and development of

glochidia in gravid females. Furthermore, glochidia may also be directly exposed to

toxicants while in situ through contact with marsupial fluids or the water column.

Additional factors that likely influence glochidia tolerances include variation in

developmental stage and maturity, genetics, and acclimation of the gravid female or

glochidia to laboratory conditions.

The individual or cumulative effects that these disturbances have on glochidia

health are unknown, as is their potential influence on test results. Lower LC50 values are

anticipated in bioassays that use glochidia collected from gravid females in a weakened

state, which may lead to false-positive test results that overestimate environmental risk.

The range of NaCl tolerances observed in our study shows that intra-specific variability is

common among glochidia obtained from different gravid females, and we expect a

similar trend in studies with other toxicants. This observation is pertinent in the

development of a standardized test protocol, and emphasizes the need for an improved

approach for assessing the health of glochidia used in bioassays.

47

2.5.2 Inter-specific variation

Glochidia from different freshwater mussel species exhibited varying NaCl

tolerances, corroborating previous studies that reported substantial differences in species

tolerances for other toxicants. Keller and Ruessler (1997) examined the toxicity of the

organophosphate insecticide malathion to glochidia from several species, and found a

wide range in tolerances. The lowest 24-h LC50 was recorded for glochidia of V. lienosa

(22 mg/L), while the highest was for U. imbecillis (366 mg/L). Similarly, Huebner and

Pynnonen (1992) reported the tolerance of glochidia from Anodonta cygnea and A.

anatina to be different in experiments testing pH and aluminum. Studies have also

shown substantial differences in species tolerances to copper, as there was a three-fold

difference in the LC50 values for the most and least sensitive species (Cherry et al. 2002,

Jacobson et al. 1997)

Few published studies have tested glochidia from multiple species, but the relative

sensitivities of species would be expected to fluctuate depending on the toxicant. For

example, a high sensitivity to NaCl is not always indicative of greater sensitivity to other

toxicants. In this study, V. iris glochidia were most sensitive to NaCl, but we found the

glochidia of this species to be far more tolerant to mercury than those of E. capsaeformis,

E. brevidens, and L. fasciola (Valenti et al. 2003). Therefore, researchers should be

cautious when attempting to use glochidia of common mussel species as surrogate test

organisms for other species in environmental risk assessments.

2.5.3 Comparison of LC50 and Control Survivorship

Glochidia of several mussel species had greater variability in LC50 values than

control survivorships. This difference was most evident in experiments with V. iris, as

48-h LC50 values varied 7-fold, while control survivorship remained >80% for glochidia

48

from the various adults. A similar trend has also been noted for C. dubia, D. magna, and

P. promelas in monthly acute NaCl reference tests, conducted according to US EPA

protocol (1993), as LC50 values among months were more variable than the control

survivorship, which was > 90% in all tests (D.S. Cherry, unpublished data). These

results suggest that the comparative evaluation of NaCl tolerances provides researchers

with a more accurate measurement of test organism quality compared to monitoring

control survivorship, because subtle differences are more readily quantifiable.

2.6 Conclusion

My study showed significant differences among the tolerances of glochidia from

different gravid adults to NaCl within (intra-specific variability) and between (inter-

specific variability) species. These findings are important because despite reported

differences in tolerances among species (Huebner and Pynnonen 1992, Jacobson et al.

1997, Keller and Ruessler 1997, Augspurger et al. 2003), no publications have examined

the variability in sensitivities of glochidia from different gravid females of the same

species. Although my study lacks sufficient results to establish acceptable ranges of

NaCl tolerances for glochidia of the species tested, it does confirm that variability in

tolerances is common within a species. The study also provides a tentative protocol for

conducting reference toxicant tests that appears acceptable for most species, common and

endangered. The incorporation of reference toxicant tests with glochidia will provide an

additional QA/QC measure that may be critical in the future development of a

standardized test protocol.

Furthermore, my study supports implementing a QA/QC criterion of > 80.0%

viability in control treatments as a validation of the health of glochidia used in bioassays,

as proposed by Keller and Ruessler (1997). The results of toxicity tests in which >20%

49

mortality occurred in control glochidia are questionable. As a result, glochidia from

some mussel species may be less effective test organisms because maintaining acceptable

levels of control survivorship may only be feasible when test durations are substantially

reduced. High levels of control mortality jeopardize test results by making it more

difficult for researchers to verify that observed mortality is related to the toxicant(s), and

not to other extraneous factors. However, as with standard test organisms, monitoring

control survivorship may not be sufficient in itself to confidently verify the health of

glochidia used in toxicity tests.

Regulatory agencies have implemented reference toxicant tests into current

protocols for standard test organisms as additional QA/QC measures. The reference

toxicant system is based on two premises: 1) healthy individuals of the same species

should have similar sensitivities to a toxicant, and 2) unhealthy organisms are more

stressed by additional burdens of strain than healthy organisms. Organisms with lower

fitness will be more sensitive in toxicity tests, and yield lower LC50 values because their

ability to tolerate a toxicant is reduced. When reference test results are compared to those

of a long-term data base, reduced organism health should become evident if reference

LC50 values are not within an established species range.

Despite > 80.0% control survivorship, markedly lower NaCl tolerances for

glochidia from some gravid females reinforces the need to improve QA/QC methods for

evaluating the health of glochidia. However, additional studies are needed to establish

adequate species data bases before QA/QC reference toxicant systems, similar to those

used for standard regulatory test organisms, can be effective. Larger sample sizes of

adults of the same species are important for establishing acceptable ranges of glochidia

tolerances. Future studies should also verify that NaCl is an acceptable reference toxicant

50

for glochidia, despite the fact that the US EPA currently uses it as a standard reference

toxicant for their freshwater organisms. Therefore, reference bioassays and other toxicant

tests need to be conducted concurrently with glochidia collected from individual females

to evaluate the relationships in tolerances If NaCl is an acceptable reference toxicant,

then glochidia obtained from females who exhibit a low tolerance to it relative to other

individuals of that species, should also have lower tolerances to other toxicants.

Reference toxicant tests may also serve as monitors for assessing the health of the

broods of gravid females collected from a population over time. As alluded to earlier, a

mussel’s ability to produce and release viable glochidia is likely associated with

environmental conditions, and may be limited by anthropogenic, biotic, and abiotic stress

factors. Monitoring the viability of glochidia from females using a reference toxicant

may become an easy, cost-effective approach for the long-term monitoring of mussel

assemblages. However, base-line reference toxicant tolerance criteria must first be

established for a given population before the system can be implemented as a biological

monitoring tool.

2.7 Acknowledgements

A federal Endangered Species Permit enabled researchers to collect specimens for

this study. Special thanks goes to Dr. Rebecca Currie, Branden Locke, and Jennie Uerz

for their assistance.

51

2.8 References

1) Augspurger T, AE Keller, MC Blacks, WG Cope, FJ Dwyer. 2003. Water quality guidance for protection of freshwater mussels (Unionidae)from ammonia exposure. Environ Toxicol Chem 22: 2569-2575.

2) Beckvar N, S Salazar, M Salazar, K Finkelstein. 2000. An in situ assessment

of mercury contamination in the Sudbury River, Massachusetts, using transplanted freshwater mussels (Elliptio complanata). Can J Fish Aquat Sci 57: 1103-1112.

3) Cherry DS, Van Hassel JH, Farris JL, Soucek DJ, Neves RJ. 2002. Site-

specific derivation of the acute copper criteria for the Clinch River, Virginia. Hum Ecol Risk Assess 8:591-601.

4) Goudrea SE, RJ Neves, RJ Sheehan. 1993. Effects of wastewater treatment

plant effluents on freshwater mollusks in the upper Clinch River, Virginia, USA. Hydrobiologia 252: 211-230.

5) Henley WF, Neves RJ. 1999. Recovery status of freshwater mussels

(Bivalvia: Unionidae) in the North Fork Holston River, Virginia. Am Malacol Bull 15:65-73.

6) Huebner JD, KS Pynnonen. 1992. Viability of glochidia of two species of

Anodonta exposed to low pH and selected metals. Can J Zool. 70: 2348-2355.

7) Jacobson PJ, Neves RJ, Cherry DS, Farris JL. 1997. Sensitivity of glochidial stages of freshwater mussels to copper. Environ Toxicol Chem 16:2384-2392.

8) Keller AE, Ruessler DS. 1997. The toxicity of malathion to unionid mussels:

Relationship to expected environmental concentrations. Environ Toxicol Chem 16:1028-1033.

9) Malley F, A Stewart, B Hall. 1996. Uptake of methyl mercury by the floater

mussel, Pyganodon grandis (Bivalvia, Unionidae), caged in a flooded wetland. Environ Toxicol Chem 15: 928-936.

10) Metcalfe-Smith J. 1994. Influence of species and sex on metal residues in

freshwater mussels (Family: Unionidae) from the St. Lawrence River, with implications for biommonitoring programs. Environ Toxicol Chem 13: 1433-1443.

11) Milam CD, Farris JF. 1998. Risk identification associated with iron-

dominated mine discharges and effect upon freshwater bivalves. Environ Toxicol Chm 17:1611-1619.

52

12) Naimo TJ. 1995. A review of the effects of heavy metals on freshwater mussels. Ecotoxicology 4:341-362.

13) Newton TJ, JW Allran, JA O’Donnell, MR Bartsch, WB Richardson. 2003.

Effects of ammonia on juvenile unionid mussels (Lampsilis cardium) in laboratory sediment toxicity tests. Environ Toxicol Chem 22: 2554-2560.

14) Salanki J, K Balogh. 1989. Physiological background for using freshwater

mussels in monitoring copper and lead pollution. Hydrobiologia 188/189: 445-454.

15) Unpublished data. Donald Cherry. Virginia Tech. Biology Department.

Blacksburg, VA. 24060.

16) US Environmental Protection Agency. 1990. Proposed guide for conducting acute toxicity tests with the early life stages of freshwater mussels. EPA/68/02/4278. Washington, DC.

17) US Environmental Protection Agency. 1993. Methods for measuring the

acute toxicity of effluents and receiving waters to freshwater and marine organisms- 4th ed. EPA/600/4-90/027F, Washington, DC.

18) US Environmental Protection Agency. 1994. Short-term methods for

estimating the chronic toxicity of effluents and receiving waters to freshwater organisms-3rd ed. EPA/600/4/91/002. Washington, DC.

19) Valenti TW, DS Cherry, RJ Neves, BA Locke. November 8-13th, 2003.

Sensitivities of freshwater mussel glochidia and regulatory test organisms to mercury and a reference toxicant. SETAC 24th Annual Meeting in North America. Austin, TX. Platform Presentation

20) Warren L, S Klaine, M Finley. 1995. Development of a field bioassay with

juvenile mussels. J N Am Benthol Soc 14: 341-346.

21) Weinstein JE, Polk KD. 2001. Phototoxicity of anthracene and pyrene to glochidia of the freshwater mussel Utterbackia imbecillis. Environ Toxicol Chem 20:2021-2028.

22) Weinstein JE. 2001. Characterization of the acute toxicity of photoactivated

fluoranthene to glochidia of the freshwater mussel, Utterbackia imbecillis. Environ Toxicol Chem 20:412-419.

23) Yeager M, D Cherry, and R Neves. 1994. Feeding and burrowing behaviors

of juvenile rainbow mussels, Villosa iris (Bivalvia: Unionidae). J N AM Benthol Soc 13: 217-222.

53

Table 2.1. Summarized test parameters for a 48-h acute reference toxicity test protocol with freshwater mussel glochidia. Test conditions are based on United States Environmental Protection Agency standard protocol USEPA/600/4-90/027F.

Test condition Static, non-renewal. Temperature 20 + 1° C. Test duration 48 h. Toxcant Chemical grade NaCl. Dilution factor 0.5. Test concentration Five test concentrations and a control. [0, 0.25, 0.5, 1.0, 2.0, 4.0 g NaCl/L]. Light quality Fluorescent lighting. Photoperiod 16:8 light:dark. Test chamber 50-ml beaker filled with ~ 40 ml of test solution. Test water Moderately hard, synthetic water (EPA100). Test organisms Glochidia obtained from gravid mussels that are extracted < 4 h prior to test initiation. Organisms/replicate 25-75 organisms per replicate. Replicates/concentration Four replicates per monitoring interval. Water quality Dissolved oxygen, pH, conductivity, hardness and alkalinity measured. Feeding None. Aeration None. Assessing viability Exposure glochidia to a concentrated NaCl solution. Any glochidia closed prior to or remaining open after the addition of the NaCl solution are to be recorded as dead. Endpoints Lethal concentration 50 (LC50) is the calculated

concentration that results in mortality for 50% of the exposed test organisms.

Test acceptability Control survivorship > 80%. Equipment Incubator, dissecting scope, and appropriate meters.

Parameter Conditions

54

Table 2.2. The LC50 values and mean survivorship for Villosa iris glochidia from eight female mussels in the control, 0.5, 1.0, and 2.0 g NaCl/L treatments.

Survivorship Time Individual Control 0.5 1.0 2.0 LC50

[95% CI]

1

97.5 A

90.5 A

92.0 A

16.5 A1.51

[1.45-1.58]

2

98.0 A

85.5 AB

72.0 B

13.5 AB1.25

[1.19-1.31]

3

97.0 A

90.4 A

74.5 AB

8.1 BC1.22

[1.15-1.29] 24

4

95.7 AB

72.6 CD

59.8 BC

7.5 BC1.06

[0.95-1.18]

5

93.5 AB

87.0 AB

74.5 A

11.0 B1.31

[1.23-1.38]

6

93.1 AB

76.7 CD

58.5 BC

4.9 CD1.06

[0.97-1.16]

7

97.5 A

74.5 CD

41.0 C

5.0 CD0.85

[0.78-0.93]

8

89.2 B

28.5 E

20.5 D

1.6 D0.31

[0.25-0.36]

1

96.0 AB

94.5 A

91.0 A

22.5 A1.60

[1.53-1.68]

2

97.0 A

85.5 AB

53.0 CD

14.0 B1.06

[0.98-1.14]

3

90.8 BC

91.8 A

74.6 AB

3.3 C1.27

[1.21-1.33] 48

4

89.6 C

75.8 BC

61.0 BC

3.4 C1.15

[1.07-1.23]

5

93.5 ABC

87.5 A

48.5 CD

0 C0.99

[0.94-1.05]

6

95.0 ABC

66.2 C

57.5 C

0 C0.97

[0.86-1.08]

7

95.0 ABC

50.0 D

37.5 D

0.5 C0.57

[0.42-0.78]

8

82.9 D

29.0 E

6.0 E

0 C< 0.25 [n/a]

Different letters represent significant differences (p<0.05)

55

Table 2.3. The LC50 values and mean survivorship for Epioblasma capsaeformis glochidia from five female mussels in the control, 0.5, 1.0, and 2.0 g NaCl/L treatments.

Different letters represent significant differences (p<0.05)

Survivorship Time Individual Control 0.5 1.0 2.0 LC50

[95% CI]

1

90.7 A

91.0 A

86.9 A

42.0 AB1.91

[1.78-2.06]

2

93.7 AB

96.3 A

91.6 A

50.9 A1.98

[1.87-2.10]

24 3

91.7 AB

93.7 A

90.5 A

36.7 B

1.81 [1.69-1.93]

4

91.0 AB

90.3 A

92.2 A

35.9 B

1.84 [1.75-1.93]

5

87.3 B

90.5 A

46.3 B

11.5 C

1.10 [1.02-1.19]

1

91.5 A

86.6 A

88.2 A

50.4 A

1.90 [1.74-2.08

2

89.1 A

87.4 A

85.5 A

24.8 B

1.65 [1.56-1.76]

48

3

80.4 B

88.9 A

84.8 A

74.4 C

1.86 [1.75-1.98]

4

76.9 B

66.5 B

70.2 B

4.2 D

1.38 [1.34-1.42]

5

n/a

n/a

n/a

n/a

n/a

56

Table 2. 4 The LC50 values and mean survivorship for Alasmidonta heterodon glochidia from two female mussels in the control, 0.5, 1.0, and 2.0 g NaCl/L treatments.

Survivorship Time Individual Control 0.5 1.0 2.0 LC50

[95% CI]

24 1

95.5 A

98.8 A

92.0 A

80.6 A

2.62 [2.51-2.74]

2

95.4 A

93.4 A

92.7 A

87.7 A

2.41 [2.25-2.58]

48

n/a

n/a

n/a

n/a

n/a

n/a

57

Table 2.5. The LC50 values and mean survivorship for Lampsilis fasciola glochidia from three female mussels in the control, 0.5, 1.0, and 2.0 g NaCl/L treatments.

Survivorship Time Individual Control 0.5 1.0 2.0 LC50

[95% CI]

1

94.3 A

91.6 A

87.0 B

57.9 A2.09

[1.96-2.22]

24 2

94.1 A

94.8 A

92.4 A

62.3 A

2.22 [2.12-2.32]

3

84.9 B

82.0 B

81.9 C

64.8 A

2.41 [2.29-2.53]

1

87.5 A

84.6 AB

80.1 B

71.8 A

2.30 [2.16-2.44]

48

2

91.3 A

90.3 A

87.6 A

50.4 B

2.00 [1.87-2.13]

3

80.0 B

82.4 B

81.2 B

48.5 B

2.06 [1.94-2.19]

58

24 hr

A . heterodon L. fasciola E . capsaeform is V illosa iris

Surv

ivor

ship

(%)

0

20

40

60

80

100

LC

50 (g

NaC

l/L)

0 .0

0.5

1.0

1.5

2.0

2.5

3.0

48 hr

Species

L. fasciola E . capsaeform is V illosa iris

Surv

ivor

ship

(%)

0

20

40

60

80

100

LC

50 (g

NaC

l/L)

0 .0

0.5

1.0

1.5

2.0

2.5

3.0

C ontrol SurvivorshipL C 50 value

Figure 2.1. The mean percent control survivorship and LC50 with corresponding standard deviation for glochidia from the mussel species Alasmidonta hererodon, Lampsilis fasciola, Epioblasma capsaeformis, and Villosa iris for a) 24-h test results, and b) 48-h test results.

59

Chapter 3: Acute and chronic sensitivity of early life stages of freshwater mussels (Bivalvia: Unionidae) to total residual chlorine relative to water quality criteria of 1984

60

3.1 Abstract

Chlorine is a highly toxic, widely used halogen disinfectant that is present in point source

pollution discharges from wastewater treatment plants and industrial facilities. Although

it dissipates quickly in natural waters, Cl pollution poses a substantial threat to biota

located downstream from these sources due to its continuous discharge. The current

United State Environmental Protection Agency freshwater criteria for Cl is 19 µg/L total

residual Cl (TRC) as a maximum 1-h average concentration, and 11 µg/L as a maximum

3-day average. Researchers are growing increasing concerned about declines not only in

the number of species and densities of individuals, but also in recruitment. Earlier life

stages of freshwater mussels are far more sensitive than adults to contaminants other than

Cl. Acute tests were conducted with glochidia from several species, while chronic

bioassays were conducted with 2-mo old Epioblasma capsaeformis and 3-, 6-, and 12-mo

old Villosa iris juveniles. The 24-h LC50 values for glochidia from V. iris, Lampsilis

fasciola, Alasmidonta heterodon, E. capsaeformis, and E. brevidens were 22, 15, 12,10,

and 7 µg TRC/L, respectively. LOAEC(s) for E. capsaeformis was 20 µg TRC/L, while

similar values for V. iris by increasing age were 30, 30, and 60 µg TRC/L. Younger age

classes were more sensitive, which was far more evident for survivorship endpoints, as

mortality in concentrations of 30-250 µg TRC/L was between 20-50% for 3-mo old, 60 –

100% for 6-mo old, and 90 – 100% for 12-mo old juveniles. Acute endpoints for

glochidia tests were 2.5 - 37 times higher than values for cladoceran species,

Ceriodaphnia dubia and Daphnia magna. Comparing chronic sensitivities for the same

species results in endpoints 3.7 – 16.2-fold lower for juveniles. Overall, the greater

sensitivity of cladocerans suggest that they may be used effectively as surrogates.

Key words: chlorine, freshwater mussels, acute, chronic, glochidia, juveniles.

61

3.2 Introduction

Chlorine is a halogen disinfectant often used by wastewater treatment facilities to

eliminate pathogenic organisms in discharges before their release into aquatic systems

(Pratt et al. 1988). It has also been used effectively as an agent to control biofouling of

exotic bivalves in waterlines of industrial and electrical plants (Cherry et al. 1986,

Doherty et al. 1986, Ramsay et al. 1988, Rajagopal et al. 1995, 1997). The high toxicity

and relative quick dissipation rate of chlorine from the water column make it an

appealing chemical alternative (Stewart et al. 1996). Because human activities cause

high concentrations of chlorine to be released into the environment, considerable research

has been conducted examining its effect on freshwater systems. Numerous studies

conducted in the late 1970s and early 1980s assessed environmental impairment at either

species or community levels (US EPA 1985). Extensive toxicity testing in the laboratory

with several early life stages of aquatic organisms, along with survey and in situ

experiments, allowed researchers to infer what concentrations would pose a threat to

biota in river systems. Based upon data from these experiments, the United States

Environmental Protection Agency (EPA) drafted the water quality criteria (WQC) for

chlorine in 1984. This document provided guidance to state regulatory agencies by

suggesting safe chlorine concentrations in freshwater and marine systems, and its overall

goal was to ensure the protection of aquatic life in waterways receiving permitted

outfalls. The maximum average one-hour concentration (acute) was not to exceed 19

µg/L more than once every three years on average, and the 4-d average (chronic)

concentrations was not to exceed 11 µg/L more than once every 3 yrs. The values were

based on 3-yr running averages because researchers assumed that it would take perturbed

systems that amount of time to recover fully if exposed to TRC above recommended

62

criteria. For comparison, chlorine concentrations in tap water are typically 1500-2000

µg/L.

These criteria were based on Total Residual Chlorine (TRC) to minimize the

complexity of monitoring chlorine since it may exist as one of several interim forms

depending upon the pH and temperature, along with the presence of organic matter and

nitrogenous compounds (Stewart et al. 1996). When placed in water, chlorine hydrolyzes

into one of two free chlorine forms (FRC), hypochlorous (HOCl) or hypochlorite ion

(OCl-). If ammonia is present, combined forms of chlorine will be formed (CRC), mono-

and dichloramine (Cairns et al. 1990, Wan et al. 2000 a, b). All four of these are highly

toxic to aquatic biota, and their total sum equals TRC. The Cl- ion by itself is relatively

benign to most freshwater organisms, and poses little toxicological threat (US EPA

1994). The predominant form of chlorine found in freshwater is monochloramine

because most streams and rivers have circum-neutral pH, along with high levels of

organic matter and ammonia, especially below wastewater treatment plants (Goudreau et

al, 1993).

Toxicity data for thirty-three freshwater species were considered in drafting the

WQC for chlorine. Acute values ranged from 28 µg TRC/L for the cladoceran Daphnia

magna to 710 µg TRC/L for the threespine stickleback (Gasterosteus aculeatus).

Chronic bioassays were conducted with only three species; however, endpoints were far

lower (3.4 – 26 µg TRC/L) with acute/chronic ratios ranging from 3.7 - 78. The most

sensitive organism was again D. magna, and confidence limits for chronic tests with this

species ranged from 2-14 µg/L (Arthur et al. 1975). However, at the time during which

experiments used to create WQC for chlorine were conducted (1958-1982), accurate

detection limits for TRC were approximately only 10 µg/L. Adverse effects have been

63

observed at chlorine concentration well below this threshold. Several studies have

revealed that algae and periphyton communities are impaired at chlorine concentrations

as low as 2 µg/L (Pratt et al. 1988, Cairns et al. 1990). Arthur et al. (1975) reported

chronic endpoints for D. magna ranging from 3.7 - 7.5 µg/L.

Additional information concerning the behavior of TRC in freshwater, coupled

with advancement of analytical methods, have allowed researchers to more accurately

assess toxicity since the drafting of the 1984 WQC. The high reactivity of compounds

that comprise TRC makes it difficult to maintain desired treatment concentrations during

bioassay, as dissipation from the water column is rapid, and varies distinctively for each

(Fisher et al. 1999, Taylor 1993). Many TRC constituents undergo photolysis, and

therefore concentrations in the water column decline rapidly when light is present. Since

bioassays are typically conducted with photoperiods that have 12-16 h of light,

substantial TRC loss may occur from the water column within hours. Deviation from

desired concentration will grow precipitously as test length increases, and to account for

the loss, test solutions must be intermittently renewed or continuously spiked with

chlorine to maintain stable treatment levels. Similar complexities of rapid loss are also

perpetuated by the use of natural waters during tests, since it often has substantial

chlorine buffering capacity due to the presence of fine particulate matter (referring to

both organic matter (FPOM) and geomorphic sediments), microscopic biota, and plant

material.

With progressive improvements in laboratory testing approaches with chlorine,

researchers have generated test endpoints that are likely more accurate estimates of

susceptibility. Many of these values are far lower than those reported in WQC, which is

attributed to exposures that were continuous rather than static or intermittent. Because

64

most permitted outfalls containing chlorinated effluents are released continuously, these

values may reflect toxicity more accurately. A study by Taylor (1993) emphasized the

importance of this in a series of experiments examining the acute tolerance of

Ceriodaphnia dubia to different forms of FRC and CRC. The researcher observed that

both forms of FRC (HOCl and OCl-) dissipated in less than 1 min in static treatments;

however, CRC concentrations were far more stable. Experiments comparing test designs

revealed the substantial affect that this may have on toxicological endpoints, as LC50

values for tests with FRC were 7-8 times lower for organisms in continuous versus static,

non-renewal exposures, whereas a difference was not evident in tests with CRC.

Twenty-four h LC50 values for C. dubia were 6 µg OCl-/L and 5 µg HOCl/L in

continuous unfed exposure tests, both well below the WQC, while static unfed values

were 48 and 35 µg/L, respectively. Feeding also substantially affected the acute toxicity,

as endpoints for these bioassays were well above those of the other experiments, as

values were 80 µg OCl-/L and 140 µg HOCl/L, which may be due to FRC binding to

available food and being lost from the water column. Additional studies have report

substantial differences in the endpoints of intermittent and continuous exposure tests with

chlorine. Fisher et al. (1999) documented far lower LC50 values for several species if

exposed during tests to treatments that were continuously spiked, rather than those

intermittently exposed, as respective values were 32 and 55 µg/L for D. magna, 78 and

301 µg/L for Hyalella azteca (amphipod), 59 and 374 µg/L for Oncorhychus mykiss

(rainbow trout), and 304 and 572 µg/L for Notemigonus crysoleucas (golden shiner).

The Clean Water Act of 1977 (Section 304 a:1) states that the EPA is to publish

WQC based on data that reflect the latest scientific knowledge; however, as criteria for

chlorine are based on data which are more than 20 yrs old, it does not appear to be the

65

case. Recently, laboratory toxicity testing with freshwater mussels has received

increased attention, as several researchers have documented that early life stages are

more susceptible to contaminants than organisms used to derive safe water concentrations

and/or assess environmental risk (Cherry et al. 2002, Augspurger et al. 2003).

Researchers are growing increasingly concerned of current population trends because

successful recruitment seems to be sparse, exemplified by the lack of young mussels,

even at sites where diverse assemblages of adults are found (Jacobson et al 1997, Henley

and Neves 1999). However, since unionids have multiple immature life stages during

reproduction, including a larval parasitic stage that requires a specific fish host, it has

proven challenging for researchers to evaluate the severity to which potential factors are

impacting recruitment. Although in situ studies and surveys are useful for identifying

impairment at specific sites, it is extremely difficult for researchers to isolate variables

and distinguish direct cause-effect relationships because of the complexity of river

systems. This is exemplified by interaction between multiple non-point and point source

stressors, competitive interaction among organisms, host-fish demographics, and varying

abiotic and biotic components. Laboratory testing allows researchers to assess the

relative tolerances of different life stages of unionids (i.e., glochidia, juvenile, adult) far

easier than could be achieved in the field.

The intent of this research was to determine whether current WQC for chlorine

are stringent enough to protect early life stages of native freshwater mussels. To achieve

this goal, a series of experiments were to be conducted with glochidia from various

species of freshwater mussels to determine their tolerance to TRC. These tests shall be

renewed intermittently so that accurate TRC concentrations can be maintained.

However, glochidia are only suitable as test organisms for acute tests, as researchers

66

report substantial declines in viability during laboratory studies after only short periods,

ranging from hours to days depending on the species (Telda and Fernando 1969, Huebner

and Pynnonen 1992, Jacobson et al. 1997). Therefore, several 21-d chronic toxicity tests

were conducted with juvenile mussels to assess their sensitivity to TRC. Bioassays

conducted with 3-, 6-, and 12-mo old Villosa iris juveniles will allow researchers to

determine which age class is most susceptible to chlorine. An additional chronic test

will also be conducted with juveniles from a species that is currently listed as federally

endangered, Epioblasma capsaeformis, to determine its tolerance relative to a species

with more stable populations.

3.3 Material and Methods

3.3.1 Test organisms

3.3.1.1 Glochidia

Gravid females of V. iris (rainbow mussel), E. capsaeformis (oyster mussel), E.

brevidens (Cumberlandian combshell), and Lampsilis fasciola (wavyrayed lampmussel)

were collected from the Clinch River, VA, whereas females of Alasmidonta heterodon

(dwarf wedgemussel) were collected from the Ashuilot River, NH. Epioblasma

capsaeformis, E. brevidens, and A. heterodon are federally endangered species.

Specimens obtained from the Clinch River were transported back to the laboratory

immediately after collection, while those from the Ashuilot River were mailed over night

in chilled coolers. Adults were acclimated to laboratory conditions in recirculating

troughs maintained at 20 ± 2°C and fed a tri-algal diet for at least 24 h prior to extraction

of glochidia. Glochidia were extracted by gently prying open a gravid female’s shell and

puncturing the marsupial tissue with a 100 cc water-filled syringe. Injecting the water

67

caused glochidia to be flushed out. The process was repeated for each gill. Glochidia

were then rinsed with clean water to remove unwanted material, and four samples of 25-

50 individuals were assessed for viability with a NaCl solution, as described by Jacobson

et al. (1997) and Goudreau et al. (1993). Viable glochidia were those that were open, and

responded to the addition of NaCl by closing, or repeatedly closing and opening their

valves. Glochidia that were closed prior to addition of the brine solution, or open but

showed no movement to the noxious substance, were recorded as ecologically dead since

it was assumed that they would be unable to attach to host fish. Only glochidia from

adults that had average viabilities of at least 90% prior to the initiation of tests were used

in experiments

3.3.1.2 Juvenile mussel

Juvenile mussels were obtained from the Virginia Polytechnic Institute and State

University Aquaculture Center, where they were produced in vivo on appropriate host

fish. The host species used for V.iris was Ambloplites rupestris (rockbass), while for E.

capsaeformis it was Cottus carolinae (banded sculpin). Infestation of host fish followed

the protocol of Zale and Neves (1982). Once fish hosts were infested with glochidia,

they were transferred to 20-l glass aquaria. Aquaria were monitored every few days to

determine when juvenile mussels began releasing from host fish. Once sufficient

numbers had dropped, juveniles were siphoned from the tank bottom and transferred to

cylindrical PVC chambers (diameter x height = 30 x 8 cm) that were sealed on one side

with a fine plastic mesh (55 microns). These chambers were held in large recirculating

troughs, and each contained a small amount of previously aerated and autoclaved

sediment. Troughs were filled with 50/50 mix of dechlorinated tap water and well water,

68

and treated daily with 30,000 cells/mL unicellular algae (Neochloris oleoabundans).

Juveniles were monitored periodically to assess health and renew sediment until they

reached their appropriate age for the bioassays. Only individuals in good health, as

evident by full guts, greenish hues, and pedal feeding, were used in bioassays.

I decided to use juveniles that were no younger than 2-mo old as test organisms

during chronic bioassays rather than newly transformed juveniles for several reasons.

Researchers have documented extremely low juvenile survivorship (<50%) 2 wk after

excystment during culturing experiments, despite high production yields immediately

following transformation (Jones and Neves 2002, Jones et al. 2004). Some researchers

speculate that this is attributable to some populations exhibiting normal bottlenecks,

which may be inevitable due to genetics. Zimmerman et al. (2003) also noted the

substantial impact that predators may have on survivorship in culturing experiments. To

account for these concerns, we used older mussels that we anticipated were beyond the

age of high mortality, and less susceptible to predators, such as flatworms. Substrate is

needed in chronic tests to meet the living requirements of juveniles, which also makes the

use of older mussels advantageous because they are far easier to relocate. However,

growth rates is also reduced with age. This makes it harder for researchers to distinguish

differences between treatments means, thereby reducing statistical power and making

predictions of NOAECs less accurate. Furthermore, Ingersoll et al. (2003) observed that

2-mo old individuals of V. iris were far more susceptible to 2- and 4-d exposures to

ammonia and chlorine than newly transformed juveniles.

3.3.2 Acute toxicity tests with glochidia

Treatments were 5, 10, 30, 60, 120, 250, and 500 µg TRC/L, plus a control.

Calcium hypochlorite (HTH-High Test Hypochlorite) was used as the toxicant and

69

moderately hard synthetic water (EPA100) described in EPA standard methods (2002),

was used as the diluent and control. Solutions for each treatment were prepared in 3-L

plastic nalgene beakers and then measured. If needed, concentrations were readjusted

using stock solutions until accurate. Concentrations were measured intermittently

throughout the test, and re-adjusted accordingly with stock solutions (approximately 2

times concentration of the treatment) to account for loss of chlorine. Concentrations

were more difficult to maintain target levels during light periods, and in higher treatments

(i.e., those > 120 µg/L).

Glochidia (approximately 25-50) were transferred with a fine-tip glass pipette to

test chambers (Fig. 3.1 [87]) that were randomly appropriated to the different treatments.

For each treatment, there were four replicates of 25-50 glochidia for each time interval

that a test was monitored. So if a test was conducted for 48 h, eight test chambers would

be loaded, whereas a test for 72 h would have twelve. As the availability of E.

capsaeformis and E. brevidens gravid females were limited, and glochidia were needed

for culturing, tests with these species were only conducted for 24 h. Three rounds of

bioassays were conducted with A. heterodon and E. capsaeformis, two rounds with L.

fasciola, and one round with V. iris and E. brevidens.

Viability was assessed after each 24 h interval, and was achieved by transferring

glochidia with fine-tip glass pipettes from test chambers to glass petri dishes for

observation with a dissecting microscope. The numbers of open and closed glochidia

were tallied, after which a concentrated NaCl solution was used to assess viability as

previous described. The computer Toxstat® Version 3.5 (West, Inc., Laramie,

Wyoming) was used to calculate LC50 values.

70

3.3.3 Chronic toxicity tests with juveniles

Chronic toxicity tests were conducted with six chlorine treatments that

approximately doubled in concentration and ranged between 5 and 500 µg TRC/L. For

the different tests, several treatment concentrations overlapped to allow for easy

comparisons. Bioassays with 2-mo old E. capsaeformis and 3-mo old V. iris had lower

concentration ranges beginning at 5 µg TRC/L, whereas tests with 6- and 12-mo V. iris

instead had a higher concentration of 500 µg/L. Treatments were prepared in 140-L re-

circulating systems (n=7) powered by 1.5 amp pumps. Each trough was filled with 120-

L of a 50/50 mix of dechlorinated tap water and reference water from a river site (Sinking

Creek, Newport, Virginia). One system served as the control, while the remaining six

were spiked with HTH to desired chlorine concentrations. Concentrations in the systems

were maintained by adding chlorine stock at a rate of ~10 L per day to compensate for

loss; un-spiked water was delivered to the control treatment. Stocks of different

concentrations were used on the treatments, and were prepared with HTH and the 50/50

water mixture. New stocks were prepared every 48 h and 20 L of water was removed

with a siphon from each trough. Systems were covered with Plexiglas to impede chlorine

loss. Treatments were maintained for a week prior to loading mussels to ensure that

concentrations were stable, and chlorine levels were monitored twice daily thereafter

In each trough were 20 individual test chambers (Fig. 3.1 [87]), each containing 2

ml of <150 micron, autoclaved, and aerated sediment from a reference site on the New

River, VA. Chambers were held in position with plastic test tube holders and oriented to

maximize uniform flow. The shell length of a juvenile was measured with the ocular lens

of a dissecting microscope, after which the individual was randomly allocated to one of

the individual test chambers (n=140). Troughs were dosed daily with 30,000 cells/ml

71

Neochloris oleoabundans. Temperature was maintained at 23 ± 1°C, and a photoperiod

of 16 light: 8 dark was established using an automatic timer.

After 21 d, juveniles were retrieved from the test chambers. The same

magnification was used to measure initial and final shell length for a given test, although

it varied among tests depending on test organism size. Survivorship was evaluated by

observing individuals at the highest magnifications (40 X). Individuals that did not

move for 2 min were recorded as dead. Movement was defined as pedal feeding, active

filtering, shell movement, or visceral mass movement observed through the shell. To

make statistical analysis more meaningful and universally conceptualized, ocular shell

lengths were converted to metric units (µm). Total growth was calculated by subtracting

initial length from final length. Proportional growth was calculated by dividing total

growth by initial length. No observable adverse effect concentrations (NOAEC) and

lowest observable adverse effect concentrations (LOAEC) were determined for growth

and survivorship based on the statistical approach described for Pimephales promelas in

standard protocol (EPA 2002) using Toxstat Version 3.5 (West, Inc., Laramie, Wyoming)

(α=0.95).

3.3.4 Water chemistry

The temperature, dissolved oxygen, pH, alkalinity, and hardness of each sample

were measured at the initiation and completion of acute tests, and biweekly for chronic

bioassays. An Accumet® (Fisher Scientific, Pittsburgh, PA, USA) pH meter with an

Accumet gel-filled combination electrode (accuracy <± 0.05 pH at 25°C) was used to

measure pH. Dissolved oxygen and conductivity were measured with Yellow Springs

model 54A meter®, and a model 30 conductivity meter®, respectively (Yellow Springs,

72

OH, USA). Total hardness and alkalinity (as mg/L CaCO3) were measured in accordance

with APHA et al (1998) through colorimetic titration.

3.5 Results

3.5.1 Acute toxicity of TRC to glochidia

Average viability was greater than 90% for glochidia of all species after 24 h in

control treatments, but was 0% at concentrations of 500 µg TRC/L (Table 3.1 [84]).

Epioblasma brevidens (LC50 = 70 µg/L), E. capsaeformis (LC50 = 107 µg/L), and A.

heterodon (LC50 = 107 µg/L) were slightly more sensitive to chlorine than L. fasciola

(LC50 = 145 µg/L), and far more sensitive than V. iris (LC50 = 220 µg/L) after 24 h of

exposure. At 250 µg TRC/L, average viability for sensitive species (<20%) was nearly

half that of the respective value for L. fasciola (35%), and less than a third of that for V.

iris (66%). In concentrations of 30 µg TRC/L and higher, viability remained greater than

90% for all species after 24 h, except E. brevidens (79-87%).

After 48 h of exposure, viability in treatments containing measurable chlorine, in

general, differed only slightly for V. iris and A. heterodon, although it decreased

substantially for L. fasciola. However, average viability of L. fasciola declined to less

than 80% in the control after 48 h, so results are suspect and may be inaccurate. Forty-

eight hour LC50 values for V. iris and A. heterodon were 260 and 95 µg/L, respectively.

Control viability remained greater than 90% for V. iris after 72 h, yet the LC50 value

(180 µg/L) was still higher than the 24 h value for the other species tested.

3.5.2 Test results for chronic bioassays with V. iris

Early life stages of V. iris juveniles were more susceptible to TRC exposure than

older stages. After 21 d of exposure, growth was significantly reduced at the same

concentration for all three age groups (Table 3.3 [86]). Juveniles of all classes in

73

treatments of 30 µg TRC/Land lower had no significant growth impairment compared to

individuals in controls (p<0.05). Average growth for individuals was reduced by 37-80%

in treatments with TRC concentration of 30-120 µg/L, and by 90% in treatments above

250 µg/L relative to the growth of juveniles in controls.

Survivorship was significantly reduced at the same concentration as growth for 3-

mo old juveniles (LOAEC = 30 µg/L), but at higher relative concentrations for 6- and 12-

mo old juveniles (LOAECs = 250 and >500 µg/L, respectively). Survivorship for 3-mo

old juveniles was less than 50% in treatments with TRC concentrations of 30 µg and

above. No 3-mo old individuals survived in the 120 µg TRC/L treatment. Six-month old

juveniles had > 85% survivorship in treatments with TRC concentrations less than 250

µg/L, 60% at 250 µg TRC/L, and 50% at 500 µg TRC/L. Few deaths were observed in

treatments with 12-mo old juveniles, as survivorship remained 80% or above in TRC

treatments up to 500 µg/L (Fig. 3.2 [88]).

3.5.3 Test results for chronic bioassay with E. capsaeformis

Two month-old E. capsaeformis juveniles were more sensitive than any age class

of V. iris. Growth was significantly reduced at concentrations of 20 µg TRC/L and

higher, as exposed individuals grew less than 20% relative to those in the control.

Growth in the control, and NOAEC treatment (10 µg/L) were 400 and 375 µm,

respectively, which differed by only ca. 6%. The numbers of observed mortalities were

also considerably high in the test, as 50% or more of the individuals perished at

concentrations of 30 µg/L and higher. All individuals in the 120 µg/L treatment died

after 21-d of exposure, whereas those in the control and 5 µg/L had average survivorship

of 80 and 100%, respectively (Table 3.3 [86]).

74

3.6 Discussion

3.6.1 Toxicity of TRC to glochidia

Glochidia from freshwater mussels that are currently listed as endangered were

more sensitive to chlorine exposure than those of species with relatively stable

populations, as 24-h LC50 values for E. capsaeformis, E. brevidens, and A. heterodon

were 50-70 % lower than those for V. iris (220 µg TRC/L). Ingersoll et al. (2004)

reported an LC50 value similar to the one reported in this study. Goudreau et al. (1993)

reported lower endpoints for V. iris (84 µg/L), which is attributable to tests being

conducted with CRC rather than TRC. Regardless, glochidia from all species tested were

far more tolerant to chlorine than most other aquatic organisms (Table 3.2 [85]). Acute

endpoints for glochidia in our study were 2.5-37 times higher than respective values for

cladoceran species (Taylor 1993, Fisher et al. 1999, EPA 1984). For comparison, Taylor

(1993) reported LC50 values of 5 µg/L hypochlorus ion (OCl-) and 6 µg/L hypochlorus

acid (HOCl-) in continuous exposure experiments. Researchers have also documented

endpoints for other genera, including several fish species and a freshwater copepod, that

are approximately half that of the values calculated for glochidia in this study.

The water quality criteria for chlorine drafted by the U.S. EPA in 1984 stated that

freshwater aquatic organisms, and their uses, would not be adversely affected by chlorine

if: 1) the maximum average 1 h concentration does not exceed 19 µg/L more than once

every three years on average, and 2) the 4-d average TRC concentration does not exceed

11 µg/L more than once every 3 yrs on average. Average viability of glochidia in

treatments containing 30 µg TRC/L was above 90% for all species except E. brevidens,

which had a respective value of 83%. Based upon these results, the risk that glochidia

will be adversely effected by concentrations in the environment seems minimal

75

Moreover, during experiments longer in duration, average viability of V. iris glochidia

after 48 and 72 h in the control and 10 µg/L treatments were nearly identical. After 48 h,

the mean viabilities in these treatments were 95 and 90%, respectively, and after 72 h,

90% and 88 µg/L, respectively. Similar trends are also apparent for data generated with

L. fasciola; however, since average viability of individuals in the control treatment was

below 80% after 48 h, test results may be unreliable. Based upon these observations, and

the fact that 24-h LC50 values are 6-20- fold higher, 11 µg TRC/L appears to be

sufficient to protect glochidia of most species. Furthermore, as the overall mean acute

values for freshwater mussels was 38 µg/L, it seems likely that surrogate organisms, such

as C. dubia and D. magna, will be effective for assessing environmental risk since they

are far more susceptible to chlorine.

When relating the results of laboratory bioassays to environmental scenarios,

there are additional factors that suggest glochidia will not be adversely affected by

permitted release of chlorine into aquatic systems. First of all, measured concentrations

of chlorine in the water column of river systems are highest just below the source, and

dissipate rapidly with increased distance downstream (Stewart et al. 1996). Therefore,

chlorine concentrations are often well below WQC in the water column of most river

reaches, whereas areas of concern are confined to short reaches below the outfall. As

glochidia were tolerant of chlorine concentrations far above the WQC, it seems unlikely

that actual concentrations in rivers will have toxic effects on glochidia. Furthermore,

glochidia are most susceptible to contaminants after being released by the gravid female,

but prior to encysting on a fish host. Consequently, the most susceptible phase is

exposed to contaminants for only brief periods. Although numerous species of glochidia

have been observed in stream drift, it is highly probable that those coming in contact

76

immediately with appropriate fish host species make up an overwhelming percentage of

individuals that successfully transform into juveniles.

3.6.2 Chronic juvenile test

The 21-d exposure tests with juveniles yielded more sensitive toxicological

endpoints than acute glochidia bioassays, as demonstrated by LOAECs of 20 – 60 µg/L

TRC compared to LC50 values of 70 – 220 µg TRC/L. However, the lowest measured

chlorine concentration in treatments that caused adverse effects was still above safe

concentrations established in the 1984 WQC for chlorine. The most sensitive test

endpoint for chlorine toxicity was derived from data for the chronic bioassays conducted

with E. capsaeformis juveniles (20 µg TRC/L). At this concentration, growth was

reduced by 85% relative to the control, whereas at the next lower concentration

(NOAEC) of 10 µg TRC/L, which was very close to the maximum allowable 4-day TRC

average (11 µg TRC/L), the difference was negligible (6%). Data generated in my study

suggest that E. capsaeformis is more sensitive than V. iris, as adverse effects were

observed for the former at lower concentrations in less time. The smaller body size of E.

capsaeformis relative to V. iris may be attribute to a difference in susceptible to

contaminants, and also accounts for the substantial difference in growth of individuals in

controls for the respective species, 400 compared to 700 µm.

Respective LOAEC values for the three age classes of V. iris show were the same

(30 µg/L). However, a difference in the sensitivity of the three age classes is far more

distinguishable if survivorship alone is considered, especially in higher treatment

concentrations (pg 88). After 21 d, only 20% of 3-mo old juveniles survived in the 120

µg TRC/L treatment, whereas all 6- and 12-mo old juveniles survived. At 250 µg

TRC/L, survival of 3-mo old juveniles was again 20%, but dropped substantially for 6-

77

mo olds at this concentration to 60%. No 12-mo old individuals died at this

concentration, and even at the highest chlorine treatment of 500 µg TRC/L, 80% of

exposed individuals survived. Similar size class effects have been observed in

experiments evaluating varying approaches for controlling biofouling by non-native

bivalves with chlorination. An extensive study by Rajagopal et al. (2003) showed that

size classes (10, 25, and 35 mm) of the marine brown mussel, Perna perna, had different

tolerances. Like my study, difference in sensitivities of the three age classes are most

striking when comparing survivorship at high concentrations. The number of hours to

reach 100% mortality for individuals exposed to 500 µg/L varied substantially among the

different classes as values for species in ascending order of size class were 84, 102, and

120 h, respectively.

During my study, I observed significant reductions in growth of individuals

exposed to measurable concentrations of TRC. Although studies examining effects of

chlorine on native freshwater mussels are rare, considerable research has been conducted

with introduced bivalves while developing biofouling treatment strategies (Cherry et al.

1986, Doherty et al. 1986, Ramsay et al. 1988, Rajagopal et al. 1995, 1997). These

experiments provide useful information pertaining to behavioral and sublethal effects

caused by chlorine exposure to bivalves, although applying toxicological endpoints to

infer relative tolerance of unionids has only limited merit because of substantial

differences between genera. Bivalves reduce periods of activity and filtration when

exposed to a noxious substance, such as Cl, and in some cases, completely avoid

exposure by sealing their valves and changing metabolism type (Naimo et al. 1995, Chen

et al. 2001). Irrevocably, the ability of bivalves to avoid exposure often makes mortality

a poor endpoint for juvenile experiments, especially for tests that are short in duration.

78

Therefore, sub-lethal effects are often far better indicators of impairment than

survivorship, and are especially important to consider for freshwater mussels since

reductions in fitness will become more evident over their long life spans. Reduced

growth is a definitive illustration of lower fitness, as juvenile mussels can neither

propagate nor survive if they do not grow. Individuals unable to obtain sufficient energy

reserves may perish due to latent mortality, especially younger juveniles that are unable

to develop sufficiently during summer-fall months to endure winter.

3.7 Conclusion

Overall, data obtained in my study strongly suggest that early life stages of

freshwater mussels are less sensitive to TRC than many other aquatic organisms. The

mean acute test endpoint for freshwater mussel glochidia in my study was 108 µg/L TRC,

which is more than twice the average value reported in the 1984 WQC for the five most

sensitive species (51 µg/L), and 4-16 times higher than more recent values reported for

cladoceran species. Chronic endpoints for juveniles were substantially lower than those

for glochidia, as values were 5 – 7 times lower for youngest age class of juvenile for a

given species. Test endpoints for all acute (70 – 220 µg TRC/L) and chronic tests (20 –

60 µg TRC/L) were above WQC for chlorine, which indicates that the toxicity threat is

only minimal. Furthermore, as adverse effects have been observed at far lower

concentrations for other species during chronic tests with test organisms commonly used

to assess environmental risk, such as C. dubia, D. magna, and P. promelas, it seems that

they may be more suitable as test organisms to assess chlorine toxicity. Furthermore, as

the sensitivities of the more typically-used test organisms to TRC were greater than early

life stages of mussels during my 39999study, it appears that they may be suitable as

surrogate test organisms to assess chlorine toxicity.

79

3.8 Acknowledgements

Research was conducted thanks to support from the United States Fish and Wildlife

Service. We thank Patrick Berry for his contributions during the initial phases of this

study.

80

3.9 Work Cited

1) Augspurger T, Keller AE, Black MC, Cope WG, Dwyer FJ. 2003. Water quality guidance for protection of freshwater mussels (Unionidae) from ammonia exposure. Environ Toxicol Chem 22: 2569-2575.

2) Cairns J, BR Niederlehner, JR Pratt. 1990. Evaluation of joint toxicity of chlorine

and ammonia to aquatic communities. Aquatic Toxicol 16:87-100. 3) Chen LY, AG Heath, RJ Neves. 2001. Comparison of oxygen consumption in

freshwater mussels (Unionidae) from different habitats during declining dissolved oxygen concentrations. Hydrobiologia 450: 209-214.

4) Cherry DS, Van Hassel JH, Farris JL, Soucek DJ, Neves RJ. 2002. Site-specific

derivation of the acute copper criteria for the Clinch River, Virginia. Hum Eco Risk Assess 8:591-601.

5) Cherry DS, RL Roy, RA Lechleitner, RA Dunhardt, GT Peters, J Cairns. 1986.

Corbicula fouling and control measures at the Celco plant, Virginia. Am Malacol Bull Special Edtion 2: 69-81.

6) Doherty FG, JL Farris, DS Cherry, J Cairns. 1986. Control of the freshwater

fouling bivalve Corbicula fluminea by halogenation. Arch Environ Contam Toxicol 15: 535-542.

7) Fisher DJ, DT Burton, LT Yonkos, SD Turley, GP Ziegler, BS Turley. 2003.

Derivation of acute ecological risk criteria for chlorite in freshwater ecosystems. Water Res 37: 4359-4368.

8) Gulley DD, West, Inc. 1996. TOXSTAT version 3.5. Fish Physiology and

Toxicology Laboratory, Department of Zoology and Physiology. University of Wyoming. Laramie, Wyoming.

9) Henley WF, RJ Neves. 1999. Recovery status of freshwater mussel (Bivalvia:

Unionidae) in the North Fork Holston River, Virginia. Am Malacol Bull 15: 65-73. 10) Huebner JD, KS Pynnonen. 1992. Viability of two species of Anodonta exposed to

low pH and selected metals. Can J Zool 70: 2348-2355. 11) Ingersoll C, N Wang, E Greer, D Whites, J Dwyer, A Roberts, T Augspurger, C

Kane, C Tibbott. 2003. Acute toxicity of copper, ammonia, and chlorine to glochidia and juveniles of freshwater mussels. Published abstract in Proceeding of Society of Environmental Toxicology and Chemistry 24th annual meeting, Austin ,TX, USA.

81

12) Jacobson PL, JL Farris, DS Cherry, RJ Neves. 1993. Juvenile freshwater mussel (Bivalvia: Unionidae) response to acute toxicity testing with copper. Environ Toxicol Chem 12:879-883.

13) Jones J, RJ Neves. 2002. Life history and propagation of the endangered fanshell

pearlymussel, Cyprogenia stegaria Rafinewque (Bivalvia: Unionidae). J N Am Benthol Soc 21: 76-88

14) Jones J, Neves RJ, Ahlstedt SA, Mair RA. 2004. Life history and propagation of

the endangered dromedary peralymussel (Dromus dromas) (Bivalvia: Unionidae). J N Am Benthol Soc 23: 515-525.

15) Pratt JR, NJ Bowers, BR Neiderlehner, J Cairns. 1988. Effects of chlorine on

microbial communities in naturally derived microcosms. Environ Toxicol Chem 7:679-687.

16) Rajagopal SV, KVK Nair, G van der Velde, HA Jenner. 1997. Response of the

mussel, Brachidontes striatulus, to chlorination: an experimental study. Aquatic Toxicol 3:135-149.

17) Rajagopal SV, P Venugopalan, KVK Nair, J. Azariah. 1995. Response of green

mussel, Perna viridis (L), to chlorine in the context of power plant biofouling control. Mar Behav Physiol 25:261-274.

18) Ramsay GG, J H Tackett, DW Morris. 1988. Effect of low-level continuous

chlorination on Corbicula fluminea - short communication. Environ Toxicol Chem 7: 855-856.

19) Stewart AJ, WR Hill, KD Ham, SW Christensen, JJ Beauchamp. 1996. Chlorine

dynamics and ambient toxicity in receiving streams. Ecol. Appl. 6:458-471. 20) Taylor, P. A. 1992. An evaluation of the toxicity of various forms of chlorine

Ceriodaphnia dubia. Environ Toxicol Chem 12:925-930. 21) Telda S, CH Fernando. 1969. Observations of Lampsilis radiata (Gmelin) infesting

yellow perch, Perca flavescens (Mitchell), in the Bay of Quinte, Lake Ontario. Can J Zool 47: 705-712.

22) U.S. Environmental Protection Agency. 1985. Ambient Water Quality Criteria for

Chlorine - 1984. Office of Water Regulations and Standards Criteria and Standards Division, Washington, DC. EPA 440/5-84-030.

23) U. S. Environmental Protection Agency. 1994. Chemical Summary for Chlorine.

EPA 749-F-94-010a.

82

24) U.S. Environmental Protection Agency. 2002. Short-term methods for estimation the chronic toxicity of effluents receiving water to freshwater organisms- 4th Ed. EPA/821-R-02-013. Washington, DC.

25) Wan MT, V Aggelen, W Cheng, C Buday. 2000a. Validation of the acute toxicity of

inorganic chloramines to the freshwater invertebrate Daphnia magna. Bull Environ Toxicol Chem 64: 213-220.

26) Wan MT, RG Watts, W Cheng. 2000b. Acute toxicity of inorganic chloramines to

Daphnia magna in two types of dilution water. Bull Environ Toxicol Chem 65: 147-152.

27) Zale AV, RJ Neves. 1992. Fish hosts of four species of lampsiline mussels

(Mollusca: Unionidae) in Big Moccasin Creek, Virginia. Can J Zool 60: 2335-2542. 28) Zimmerman LL, Neves RJ, Smith DG. 2003. Control of predacious flatworms

Macrostomum sp. in culturing juvenile freshwater mussels. N Amer J Aquatic 65: 28-32.

83

Table 3.1. The average viability of glochidia from five species of freshwater mussels exposed to total residual chlorine (TRC).

Average Viability (% ± SD)

Species Concentation (µg TRC/L) nb 24- h nb 48-h nb 72-h)

Controla 4 95 ± 1 4 95 ± 2 4 90 ± 4 5 4 96 ± 3 4 93 ± 3 4 87 ± 4 10 4 94 ± 2 4 90 ± 6 4 88 ± 5

Villosa iris 30 4 90 ± 7 4 95 ± 2 4 80 ± 6 60 4 91 ± 4 4 89 ± 6 4 75 ± 7 120 4 71 ± 10 4 80 ± 7 4 67 ± 17 250 4 66 ± 7 4 69 ± 7 4 49 ± 10 500 4 0 ± 0 4 0 ± 0 4 0 ± 0 Controla 8 93 ± 3 8 79 ± 9 4 55 ± 4 5 8 94 ± 4 8 78 ± 10 4 54 ± 8 10 8 93 ± 4 8 78 ± 6 4 48 ± 9

Lampsilis fasciola 30 8 91 ± 4 8 72 ± 9 4 48 ± 11 60 8 86 ± 4 8 52 ± 19 4 42 ± 13 120 8 49 ± 20 8 27 ± 13 4 24 ± 10 250 8 35 ± 13 8 4 ± 4 4 5 ± 4 500 8 0 ± 0 8 0 ± 0 4 0 ± 0 Controla 12 93 ± 6 8 93 ± 4 5 4 93 ± 4 na na 10 12 91 ± 4 8 90 ± 6

Alasmidonta heterodon 30 12 90 ± 6 8 87 ± 7 na 60 12 67 ± 10 8 58 ± 12 120 12 44 ± 11 8 35 ± 11 250 12 19 ± 13 8 11 ± 9 500 12 0 ± 0 8 0 ± 0 Controla 12 94 ± 4 5 12 93 ± 3 10 12 95 ± 3

Epioblasma capsaeformis 30 12 90 ± 3 na na 60 12 73 ± 7 120 12 42 ± 9 250 12 12 ± 8 500 12 0 ± 0 Controla 4 91 ± 7 5 4 87 ± 11 10 4 79 ± 11

Epioblasma brevidens 30 4 83 ± 14 na na 60 4 61 ± 11 120 4 22 ± 10 250 4 4 ± 5 500 4 0 ± 0

a = Concentration below detection limit. b = Number of replicates, each containing 25 or 50 individual glochidia.

84

Table 3.2. Comparison of acute toxicological endpoints for common freshwater test organisms, and those for freshwater mussel glochidia generated in our study.

Typical regulatory test organisms Freshwater mussel glochidia

Species Mean Acute Value

(µg/L) Species Time (h)

Mean EC50 value (µg/L)

Cladoceran 6 – 27a Rainbow mussel 24 220 Ceriodaphnia dubia < 80b Villosa iris 48 260

120E Cladoceran 28c 72 180

Daphnia magna 32d Wavyrayed lampmussel 24 145 Pugnose shiner 45c Lampsilis fasciola 48 80*

Notropis arogenus 72 90* Common shiner 51c Oyster mussel 24 107

Notropis cornutus Epioblasma capsaeformis Lake trout

60c Cumberland combshell 24 70 Salvelinus namaycush

Epioblasma brevidens Rainbow trout

62c Dwarf wedgemussel 24 107 Salmo gairdneri 59d Alasmidonta heterodon 48 95

Copepod 63c Epischura lacustris

Amphipod 78d Hyalella azteca * Control survivorship below 80%. a = Taylor PA (1993). Values are for concentrations of free residual or combined forms of chlorine rather than TRC. b = Stewart et al. (1996) c = Fisher et al. (1999) d = 1984 Chlorine water quality criteria. e = Manning et al. 1996.

85

Table 3.3. Initial length and growth of mussel juveniles exposed to different concentrations of Total Residual Chlorine (TRC) for 21 days.

a = No observable adverse effects concentration. b = Lowest observable adverse effects concentration.

* Significantly different that the control treatment (p<0.005).

Species Age Concentration (µg TRC/L)

Initial length(µm)

Final length (µm)

Total growth (µm)

Control 500 900 400 5 530 930 400

Epioblasma 10a 490 860 370 capsaeformis 2-mo 20b 510 580 70*

30 510 550 40* 60 520 590 70* 120 500 500 0* Control 1290 1970 680 5 1230 1900 670 15 1300 1850 550 3-mo. 30a 1330 1750 420 60b 1340 1360 20* 120 1390 1490 100* 250 1190 1260 70* Control 1900 2580 680 15 1660 2040 380 30a 1840 1980 140

Villosa iris 6-mo. 60b 1840 1950 110* 120 1730 1880 150* 250 1690 1730 40* 500 1820 1830 10* Control 6630 7980 1350 15 6550 7790 1240 30a 6920 7380 460 12-mo. 60b 6580 6690 110* 120 6870 7110 240* 250 6330 6340 10* 500 7230 7280 50*

86

(a) (b)Figure depicts (a) external view of chamber and (b) internal view of chamber with glochidia or sediment and test organisms enclosed.

Test chamber constructed of rigid

plastic tubing

(Height= 14 cm, OD= 2.5 cm)

Nitex mesh covering flow through opening

(Juveniles: 200 microns) (Glochidia: 50 microns)

One juvenile or 25-50 glochidia per test

chamber

Specified layer of sediment (<150 microns) for chronic

juvenile tests only

Figure 3.1. Diagram of test chamber used for acute glochidia and chronic juvenile toxicity tests. Not drawn to scale.

87

0

20

40

60

80

100

0 15 30 60 120 250

TRC concentration (ppb)

Ave

rage

sur

vivo

rshi

p

3-mo6-mo12-mo

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0 50 100 150 200 250

TRC concentreation (ppb)

Perc

ent g

row

th

3-mo

6-mo

12-mo

Figure 3.2. Survivorship and proportional growth for 2-, 6-, and 12-mo old Villosa iris exposed to different concentration of TRC.

88

Chapter 4: Acute and chronic toxicity of mercury to early life stages of the rainbow

mussel, Villosa iris (Bivalvia: Unionidae) -Short Communication.

89

4.1 Abstract

Mercury (Hg) contamination is receiving increased attention globally due to human

health and environmental concerns. Few laboratory studies have examined the toxicity of

Hg on early life stages of freshwater mussels, despite evidence that glochidia and juvenile

life stages are more sensitive to contaminants than adults. Three bioassays (72-h acute

glochidia, 96-h acute juvenile, and 21-day chronic juvenile toxicity tests) were conducted

by exposing Villosa iris to mercuric chloride salt (HgCl2). Glochidia were more sensitive

to acute exposure than juvenile mussels, as 24-, 48, and 72-h LC50 values for glochidia

were >107, 39, and 14 µg Hg/L, respectively. The 24-, 48-, 72-, and 96-h values for

juveniles were 162, 135, 114, and 99 µg Hg/L, respectively. In the chronic test,

juveniles exposed to Hg treatments > 8 µg/L grew significantly less than control

organisms. The substantial difference in juvenile test endpoints emphasizes the

importance of assessing chronic exposure and sub-lethal effects. Overall, my study

supports the use of glochidia as an alternative life stage to juveniles in acute toxicity tests.

However, as glochidia may only be used only in short-term tests, it is imperative that a

integrated approach be taken when assessing risk to freshwater mussels, as their unique

life history is atypical of standard test organisms. Therefore, we strongly advocate the

use of both glochidia and juvenile life stages for risk assessment.

Keywords: Freshwater mussel / Mercury / Glochidia / Juvenile / Chronic

90

4.2 Introduction

As scientists become more aware of risks that mercury (Hg) poses to humans and

wildlife, concerns for its effects on aquatic ecosystems continue to heighten. Freshwater

mussels are currently one of the fastest declining faunal groups in North America, and

may be more susceptible to Hg pollution than other aquatic organisms. Mussel

assemblages are often congregated in depositional zones (Yeager et al. 1994), and these

areas likely have higher Hg concentrations due to its affinity to bind with fine particulate

matter (Cattani et al. 1999, French et al. 1999, Wiener and Shields 2000). Several in situ

studies have shown that bivalves have a propensity to bioaccumulate Hg (Cattani et al.

1999, Wiener and Shields 2000), which may be due to their close association with the

water column- sediment boundary, and their feeding behavior(Salanki and Balogh 1989,

Naimo 1995). The United States Environmental Protection Agency (US EPA) is

currently reviewing the Water Quality Criteria (WQC) for Hg, but only limited

toxicological data are available for most aquatic species. Furthermore, few studies have

examined the sensitivities of early life stages of freshwater mussels.

Conducting bioassays with freshwater mussels in the laboratory is critical to their

conservation because it will enable researchers to determine toxicity under controlled

conditions that are not achievable in the field. Previous studies report that early life

stages of mussels are more sensitive to contaminants than adults (Yeager et al. 1994,

Naimo 1995, Jacobson et al. 1997). This finding is also supported by field observations,

as alarmingly few young mussels have been found in assemblages with diverse adult

populations (Jacobson et al. 1997, Weinstein 2001). Furthermore, immature stages of

unionids have been documented to be more sensitive than other aquatic species

(Goudreau et al. 1993, Weinstein and Polk 2001), including commonly used regulatory

91

test organisms (Cherry et al. 2002, Valenti et al. In review). However, with standard test

protocols not yet established for freshwater mussels, regulatory agencies remain hesitant

to apply test results to policy decisions.

Procedures for acute bioassays are better established for glochidia than juveniles,

as the former have been more available for testing since they are obtained from gravid

females collected from the field rather than cultured. However, limitations with using

glochidia as test organisms have become evident with researchers reporting substantial

declines in viability during laboratory studies after only short periods, ranging from hours

to days depending on the species (Tedla and Fernando 1969, Huebner and Pynnonen

1992, Jacobson et al. 1997). This may be attributable to their limited energy reserves, and

therefore glochidia are only effective as test organisms for assessing acute. Although test

duration is less limited for juvenile bioassays, there have been problems associated with

their use as test organisms in chronic tests. Researchers have had difficulty trying to

determine test approaches that meets the unique living requirements of juvenile mussels,

and as a result, there is little published literature documenting their chronic sensitivities

(Keller et al. 1999).

The purpose of this study was to conduct several toxicity tests in the laboratory

with early life stages of the rainbow mussel (Villosa iris) to determine their sensitivity to

Hg. Villosa iris was selected as the test organism because it is a widespread species in

the Southern Appalachians that has been successfully reared in the laboratory. We

conducted acute bioassays with glochidia and 2-mo old juveniles, and a 21-day, chronic

test with juveniles. Younger juveniles are likely more susceptible to contaminants, but

we chose not to use them because of the population bottleneck exhibited by many

species. Researchers have documented extremely low juvenile survivorship 2 wk after

92

successful transformation (<50%) (Jones and Neves 2002, Jones et al. 2004), which may

be attributable to high predation (Zimmerman et al. 2003). Therefore, we decided to

conduct bioassays with older, more developed juveniles that are past the phase when high

mortality is still anticipated in the laboratory since concerns about control survivorship

have brought the validity of previous unionid tests into question.

4.3 Methods

4.3.1 Test organisms

Gravid V. iris females were collected from the Clinch River, VA, USA, and

transported in water-filled coolers to the laboratory where they were acclimated for 48 h

in re-circulating troughs (20°C). To extract glochidia, the valves of an adult female were

pried open and a syringe filled with water from the troughs was inserted into the

marsupial gill. The water was then slowly injected, thus causing the glochidia in the

swollen gill tissue to be flushed and released. After glochidia were extracted, four

samples from each adult were assessed for viability using a concentrated NaCl solution,

as described by Jacobson et al. (1997) and Goudreau et al. (1993). Glochidia from three

adult females with >90% initial viability were combined and used in the toxicity test.

Juvenile mussels were produced in vivo using rock bass (Ambloplites rupestris) as

fish hosts at the Virginia Tech Aquaculture Center (Blacksburg, VA, USA). Fish hosts

were infested according to the procedure described by Zale and Neves [19]. After

juvenile mussels excysted from fish hosts, they were reared for ~ 2 mo before being used

as test organisms. Over this period, mussels were housed in sediment (150 micron)

within recirculating troughs, and fed a daily diet of 3x107 cells Neochloris oleoabundans

algae/ L.

93

4.3.2 Acute toxicity tests

4.3.2.1 Glochidia

Test conditions are summarized in Table 4.1 [105]. Viability of glochidia was

determined by transferring a sub-sample (~50) of glochidia from a replicate to a glass

petri dish for observation under magnification (40x). The total number of glochidia, and

number of closed glochidia, were tallied, after which a concentrated salt solution (20 g

NaCl/L) was added. Then the number of glochidia not responding to NaCl by

contracting their valves was recorded. Any glochidia closed prior to, or open but not

responding to NaCl, were classified as functionally dead based on the premise that they

would be unable to attach to host fish (Goudreau et al. 1993, Jacobson et al. 1997).

4.3.2.2 Juveniles

Test conditions are summarized in Table 4.1 [105]. Juvenile mussels were

randomly appropriated to replicates by transferring them with a fine tip glass pipette. To

determine survival, mussels in each replicate were observed under magnification (40x)

for movement (defined as pedal feeding, active filtering, valve contraction(s), or visceral

mass movement observed through the shell). Individuals showing no movement for 3

min were recorded as dead.

4.3.3 Chronic toxicity test with juveniles

Test conditions are summarized in Table 4.1[105]. The test apparatus was a

modified version of the self-contained simulated lotic microcosm (SLM) described by

Kennedy et al. (2003). that provides flow for lotic organisms. Each SLM consisted of

five small, glass vials (outside diameter x height = 28 mm x 15 mm) placed in a glass

petri dish housed in a 1-L beaker filled with 950 ml of test solution. The petri dish rested

94

on top of two inverted, 50-ml glass beakers. Each vial was filled with approximately 2

ml of sediment sieved to <200 µm, and held one juvenile (n = 20). A 1-ml glass pipette

connected to an air source was placed into the test apparatus.

A juvenile was randomly selected and measured under magnification (40x) using

an ocular micrometer lens before being transferred to test chambers. Shell length was

converted to mm. Treatment water was renewed every third day by siphoning and

replacing 50% with fresh test solution. The test chambers were supplied daily with

30,000 Neochloris oleoabundans cells /ml as food. Test organisms were removed after

21 d, assessed for survivorship, and measured for length, as previously described.

Mussels were found by rinsing the contents of a vial into a 250-µm sieve, which caught

the juveniles but allowed sediment to pass through.

4.3.4 Mercury analysis

Samples were prepared for Hg analysis through Inductively Coupled Plasma

(ICP) spectrometry at Virginia Tech’s soil testing laboratory, according to US EPA

standard methods (1991). For the glochidia bioassay, initial treatment Hg concentrations

were measured at time 0 h by preparing a sample of water used to fill test replicates. At

24, 48, and 72 h, samples of out-water from replicates for each treatment were combined

and analyzed. Mercury concentrations for the acute juvenile bioassay were measured at

time 0 and 96 h, except for the highest treatment, which was measured at 24 h due to

complete mortality. During the chronic juvenile test, treatment concentrations were

measured for in-water (days 1, 4, 7, 10, 13, 17, 20) and out-water (days 4, 7, 10, 13, 17,

20, 21). The same in-water was used to fill all replicates of a given treatment, and out-

water samples from each replicate were combined before being analyzed.

95

4.3.5 Statistical analysis

The Toxstat® Version 3.5 (West, Inc., Laramie, Wyoming) (Gulley 1996)

computer program was used to calculate trimmed Spearman-Karber LC50 values for the

acute bioassays, and No Observable Adverse Effects Concentration (NOAEC) and

Lowest Observable Adverse Effects Concentration (LOAEC) for the chronic test

(p=0.05). Data analysis for survivorship and growth followed US EPA protocol (2002)

for chronic bioassays with Pimephales promelas. Growth (mm) was calculated by

subtracting initial length from final length. Measured Hg concentrations were used when

values were not below detection limit.

4.4 Results and Discussion

4.4.1 Acute glochidia and juvenile toxicity tests

Villosa iris glochidia were more sensitive than 2 mo-old juveniles to acute Hg

exposure, as there was nearly a 10-fold difference in 72-h LC50 values. In both tests,

there was a dose-dependant response, and toxicity increased with exposure time. After

24 h, glochidia viability remained high, and a LC50 value could not be calculated

because survivorship was >50% in the highest test treatment (107 µg Hg/L, Table 4.2

[106]). The viability of glochidia decreased substantial after 72 h, and was <31% for

treatments >12 µg Hg/L. Viability in the control and lowest test treatment remained high

throughout both tests, >90% and >85%, respectively. During the acute test with

juveniles, all individuals died in the highest test concentration after 24 h (234 µg Hg/L<

Table 4.3 [107]), but survivorship remained high for the other treatments (>90%, pg 107).

Survivorship decreased slightly at each time interval the test was monitored, but

remained >95% after 96 h in treatments >26 µg Hg/L.

96

There have been few studies that have examined the toxicity of Hg to early life

stages of freshwater mussels. Valenti et al. (in review) reported 48-h LC50 values

spanning < 8 to 43 µg/L Hg/L for glochidia of several species, with V. iris being the most

tolerant species. Keller and Zam (1991) conducted bioassays with newly transformed

Anodonta imbecilis juveniles (1-2 day old), and reported 48- and 96-h LC50 values of

233 µg/L and 171 µg/L, respectively. In experiments similar to those of this study with

3-mo old endangered juvenile oyster mussels (Epioblasma capsaeformis), I recorded 24-

h and 48-h LC50 values of 160 and 140 µg/L total Hg, respectively. Toxicological

endpoints similar to those calculated in my study have been generated in studies with

juvenile marine bivalves, as LC50 values for different species were between 125-161 µg

Hg/L (Nelson et al. 1988, Krishnakumar et al. 1989).

Although laboratory toxicity testing with early life stages of freshwater mussels

has become a more often utilized approach for assessing environmental risk, few

researchers have compared the toxicity of a contaminant(s) to glochidia and juveniles of

the same species. Jacobson et al. (1997) compared the copper sensitivity for different life

stages of two species of freshwater mussels. In their study, acute endpoints for V. iris

released glochidia and juvenile mussels were 36-80 µg/L and 83 µg/L, respectively, while

those for Pyganodon grandis were 46-347 µg/L and 33-44 µg/L. Augspurger et al.

(2003) documented lower mean LC50 values for glochidia than juveniles of three species

(Actinonaias pectorosa, Utterbackia imbecillis, and V. iris) despite shorter exposure

times to ammonia. Glochidia from three species of freshwater mussels (U. imbecillis, V.

lienosa, and V. villosa) were also found to be more sensitive than juveniles in studies

conducted by Keller and Ruessler (1997) that examined the toxicity of malathion.

97

The results of previous studies, in addition to those of our experiments, suggest

that glochidia may be a more appropriate life stage for assessing acute toxicity based on

the regulatory approach of deriving risk from the most sensitive life stage of a species.

Glochidia are advantageous as test organisms because they are easily obtained, and more

cost effective to use in toxicological studies than juveniles. Sufficient numbers can be

obtained from only a few females, which is especially important when assessing risk for

endangered species, or species that have yet been successfully cultured in the laboratory.

Furthermore, high viability has been observed in control treatments (>90%) after 48 h

repeatedly in experiments with glochidia from V. iris and other species (Valenti et al.

2003). In addition, glochidia, unlike juveniles, are less able to avoid toxicants because

they have thinner, more permeable shells (Hoggarth and Gaunt 1988). Conversely,

juvenile mussels that are several months old may be able to avoid toxicants by altering

their normal metabolism to a lipid catabolism. This would enable them to reduce their

filtration rates and close their valves for extended periods to avoid contaminants, which

would limit their effectiveness as test organisms.

4.4.2 Chronic 21-day juvenile toxicity test

The mean growth of individuals in the control and 4 µg Hg/L treatments were

>0.5 mm, and did not vary significantly (Table 4.4 [108]). Growth was reduced by 25%

at 8 µg Hg/L, and by >50% in the remaining treatments. Individuals exposed to >32 µg

Hg/L grew only ca. 10% as much as the control. No dead juveniles were found, and the

only apparent mortalities recorded were for individuals not located. NOAEC and

LOAEC for growth were 4 and 8 µg Hg/L, respectively.

Several studies have shown that bivalves decrease oxygen consumption, growth,

and byssal thread production when exposed to Hg (Nelson et al. 1988, Naimo et al. 1995,

98

Jacobson et al. 1997). Salanki and Balogh (1989) reported that Hg affects filtration rates

of bivalves, as exposed individuals had shorter periods of activity, and extended periods

of rest. The lower growth observed in treatments containing higher Hg concentrations

may be due to ingesting less food for assimilation into new body tissue. The high

survivorship despite the lack of growth observed in the chronic bioassay emphasizes the

importance of assessing sub-lethal effects. Although these types of impairment do not

cause immediate mortality, they likely have adverse latent effects on survivorship, and

may be more appropriate endpoints for assessing environmental impairment.

The absence of mortality in the chronic bioassay is attributable to several

differences between the acute and chronic test designs. The addition of sediment in the

chronic tests likely reduced toxicity by serving as a physical barrier to exposure, or by

binding up some of the Hg. Keller et al. (1999) reported that the use of silt in juvenile

mussel experiments removed a substantial portion of copper from the water column.

Difficulties with maintaining constant Hg concentrations in the water column caused

juveniles to only be intermittently, rather than chronically exposed to Hg. Furthermore,

not feeding mussels in the acute test also may have expedited their uptake rate, as Naimo

(1995) noted that some bivalves accumulate Hg faster when not fed. Acute bioassay test

conditions may have stressed juveniles, causing them to be more susceptible to Hg,

because test chambers were not aerated and did not contain sediment. In early attempted

experiments, high mortality was observed (>50%) in control treatments without sediment

or aeration after 7 d, despite feeding and water renewal.

4.5 Conclusions

My study supports the use of glochidia as an alternative life stage for juveniles in

acute bioassays since their ability to avoid exposure is limited, and they are more readily

99

available and cost-effective to obtain for toxicity tests. There is concern that conducting

acute bioassays with older juveniles could lead researchers to underestimate toxicity,

since individuals may be able to avoid toxicants by closing their valves for sustained

periods. However, there are limitations with using glochidia as test organisms, since they

may be used only in short-term tests. Therefore it is imperative that a new integrated

approach be taken when assessing risk to freshwater mussels, as their unique life history

is atypical of standard test organisms. I strongly advocate the use of both glochidia and

juvenile life stages in future toxicological studies. In addition to standard test organisms,

glochidia may be ideally used during initial phases of risk assessment in acute

experiments. These tests will provide researchers with a cost-effective means for

determining whether further investigation is needed for specific environmental scenarios.

If reasonable threat is apparent, then chronic tests with juvenile mussels may be

warranted.

Overall, it is important that researchers be cautioous when using findings from

laboratory bioassays to infer environmental risk. Glochidia are most susceptible to

contaminants in the water column after being released by the gravid female and before

encysting on host fish (Jacobson et al. 1997), which is often only a short time interval.

Conversely, juvenile mussels are more likely threatened by chronic exposure to toxicants

in sediment or interstitial water, as Yeager et al. (1994) noted that juvenile V. iris burrow

into the substrate and rely on pedal feeding, rather than filter feeding. Additional

research exploring new testing techniques is needed before researchers are able to

determine workable approaches for assessing pollution risk to mussel assemblages.

100

4.6 Acknowledgements

I thank the United States Fish and Wildlife Service for their support of this

project. I also thank Jess Jones and Rachel Mair of the Virginia Tech Aquaculture Center

for their assistance in obtaining and culturing test organisms.

101

4.7 References

1) Yeager MM, Cherry DS, Neves RJ. 1994. Feeding and burrowing behavior of juvenile rainbow mussels, Villosa iris (Bivalvia: Unionidae). J North Am Benthol Soc 13:217-222.

2) Cattani O, Fabbri D, Salvati M, Trombini C, Vassura I. 1999. Biomonitoring of

mercury pollution in wetland near Ravenna, Italy by translocated bivalves (Mytilus galloprovincialis). Environ Toxicol Chem 18: 1801-1805.

3) French KJ, Scruton DA, Anderson MR, Schneider DC. 1999. Influence of

physical and chemical characteristics on mercury in aquatic sediment. Water Air Soil Poll 110: 347-362.

4) Wiener JG, Sheilds PJ. 2000. Mercury in the Sudbury River (Massachusetts,

USA): pollution history and synthesis of recent research. Can J Fish Aquat Sci. 57: 1053-1061.

5) Salanki J, Balogh KV. 1989. Physiological background for using freshwater

mussels in monitoring copper and lead pollution. Hydrobiologia 188/189: 445-454.

6) Naimo TJ. 1995. A review of the effects of heavy metals on freshwater mussels.

Ecotoxicology 4:341-362.

7) Jacobson PJ, Neves RJ, Cherry DS, Farris JL. 1997. Sensitivity of glochidial stages of freshwater mussels (Bivalvia: Unionidae) to copper. Environ Toxicol Chem 16:2384-2392.

8) Weinstein JE. 2001. Characterization of the acute toxicity of photoactivated

fluoranthene to glochidia of the freshwater mussel Utterbackia imbecillis. Environ Toxicol Chem 20:412-419.

9) Goudrea SE, Neves RJ, Sheehan RJ. 1993. Effects of wastewater treatment plant

effluents on freshwater mollusks in the upper Clinch River, Virginia, USA. Hydrobiologia 252: 211-230.

10) Weinstein JE, Polk KD. 2001. Phototoxicity of anthracene and pyrene to

glochidia of the freshwater mussel Utterbackia imbecillis. Environ Toxicol Chem 20:2021-2028.

11) Cherry DS, Van Hassel JH, Farris JL, Soucek DJ, Neves RJ. 2002. Site-specific

derivation of the acute copper criteria for the Clinch River, Virginia. Human Ecol Risk Assess 8:591-601.

102

12) Valenti TW, Cherry DS, Neves RJ, Locke BA. (In review). Bivalve Ecotoxicology. eds. Van Hassel and Farris. Sensitivity of mussel glochidia and regulatory test organisms to mercury and a reference toxicant. Presented at SETAC.

13) Tedla S, Fernando CH. 1969. Observations of Lampsilis radiata (Gmelin)

infesting yellow perch, Perca flavescens (Mitchell), in the Bay of Quinte, Lake Ontario. Can J Zool 47: 705-712.

14) Huebner JD, Pynnonen KS. 1992. Viability of glochidia of two species of

Anondonta exposed to low pH and selected metals. Can J Zool 70: 2348-2355.

15) Keller AE, Ruessler DS, Kernaghan NJ. 1999. Effect of test conditions on the toxicity of copper to juvenile unionid mussels. In Environmental Toxicology and Risk Assessment: Standardization of Biomarkers for Endocrine Disruption, Vol 8. STP 1364. American Society for testing and Materials, Philadelphia, PA, pp 329-340.

16) Jones J, Neves RJ. 2002. Life history and propagation of the endangered fanshell

pearlymussel, Cyprogenia stegaria Rafinesque (Bivalvia: Unionidae) J North Am Benthol Soc 21: 76-88.

17) Jones J, Neves RJ, Ahlstedt SA, Mair RA. 2004. Life history and propagation of

the endangered dromedary peralymussel (Dromus dromas) (Bivalvia: Unionidae). J N Am Benthol Soc 23: 515-525.

18) Zimmerman LL, Neves RJ, Smith DG. 2003. Control of predacious flatworms

Macrostomum sp. in culturing juvenile freshwater mussels. N Am J Aquatic 65: 28-32.

19) Zale AV, Neves RJ. 1982. Fish host of four species of Lampsiline mussels

(Mollusca: Unionidae) in Big Moccasin Creek, VA. Can J Zool 60: 2535-2542.

20) Kennedy AJ, Cherry DS, Currie RJ. 2003. Evaluation of ecologically relevant bioassays for a lotic system impacted by a coal-mining effluent using Isonychia biocolor. Environ Monit Assess 44: 324-331.

21) U.S. Environmental Protection Agency. 1991. Methods for determining metals

in environmental samples. EPA/600/4-91/010. Washington, DC.

22) Gulley DD, West, Inc. 1996. TOXSTAT version 3.5. Fish Physiology and Toxicology Laboratory, Department of Zoology and Physiology. University of Wyoming. Laramie, Wyoming.

23) U.S. Environmental Protection Agency. 2002. Short-term methods for estimation

the chronic toxicity of effluents receiving water to freshwater organisms- 4th Ed. EPA/821-R-02-013. Washington, DC.

103

24) Keller AE, Zam SG. 1991. The acute toxicity of selected metals to the freshwater mussel, Anodonta imbecillis. Environ Toxicol Chem 10:539-546.

25) Nelson DA, Miller JE, Calabrese A. 1988. Effect of heavy metals on bay

scallops, surf clams, and blue mussels in acute and long-term exposures. Arch Environ Contam Tox 17: 595–600.

26) Krishnakumar PK, Damondaran R, Nambisan PNK. Acute toxicity of selected

heavy metals to the green mussel Perna viridis. Indian J Marine Sci 16:263-264.

27) Augspurger T, Keller AE, Black MC, Cope WG, Dwyer FJ. 2003. Water quality guidance for protection of freshwater mussels (Unionidae) from ammonia exposure. Environ Toxicol Chem 22: 2569-2575.

28) Keller AE, Ruessler DS. 1997. The toxicity of malathion to unionid mussels:

Relationship to expected environmental concentration. Environ Toxicol Chem 16: 1028-1033.

29) Hoggarth MA, Gaunt AS. 1988. Mechanics of glochidial attachment (Mollusca:

Bivalvia: Unionidae). J Morphol 198: 71-81.

104

105

Table 4.1. Summary of test parameters for acute tests with glochidia and juveniles, and the chronic test with juvenile freshwater mussels.

•

•

TEST TYPE AND LIFE STAGE

PARAMETERS Acute-Glochidia Acute-Juvenile Chronic-Juvenile

Test type Static / non-renewal Static / non-renewal Dynamic / renewal

Temperature 20 + 1 º C 20 + 1 º C 20 + 1 º C

Test duration 72 h 96 h 21 days

Toxicant Mercuric chloride (MC) MC MC

Test concentrationsA 0, 8, 15, 30, 60 120 µg/L 0, 15, 30, 60 120, 250 µg/L 0, 4, 8, 15, 30, 60 120 µg/L

Test chamber 50-ml glass beaker 50-ml glass beaker 1-L glass beaker

Test solution 40 ml 40 ml 950 ml

Diluent waterB EPA100 EPA100 EPA100

Sediment none none 2 ml sieved to 200 microns

Test organisms < 2 h after extraction 2 months old 2 months old

Organisms/replicate 50 – 75 5 5

# of Replicates 4 per time interval 4 total 4 total

Feeding none none 30,000 cells daily

Aeration none none yes

Endpoints

LC50

LC50

Survivorship and Growth

NOAEC + LOAEC

A= Nominal concentrations. B= Moderate hard synthetic water, prepared according to standard US EPA protocol EPA-821-R-02-012.

106

Table 4.2. Mean survivorship and LC50 values after 24, 48, and 72 h for Villosa iris glochidia exposed to different concentrations of mercuric chloride (HgCl2).

A= Below detection limit (<8.4 µg/L) and expressed as nominal value. B= To calculated mean concentration, 8.4 µg/L was used as the 72-h value since actual concentration was below detection limit. C= Insufficient mortality to generate an LC50 value.

Total [Hg](µg/L)

24 h Survivorship

(%)

48 h Survivorship

(%)

72 h Survivorship

(%) ControlA 94 95 91

8A 96 95 85

12B 94 92 31

25.5 94 86 0

62 92 10 0

107 62 0 0

LC50 Value > 107C 39 14

95% Confidence Limits na 37-41 12-15

Table 4.3. Mean survivorship and LC50 values after 24, 48, 72, and 96 h for Villosa iris juveniles exposed to different concentrations of mercuric chloride (HgCl2).

A= Below detection limit (<8.4 µg/L) and expressed as nominal value.

B= To calculated mean concentration, 8.4 µg/L was used as the 72-h value since actual concentration was below detection limit.

Total [Hg]

(µg/L)

24 h Survivorship

(%)

48 h Survivorship

(%)

72 h Survivorship

(%)

96 h Survivorship

(%) 0A 100 100 100 95

13B 100

100 95 95

26 100 100 100 95

59 100 90 85 75

129 90 75 55 40

234 0 0 0 0

LC50 Value 162 135 114 99

95% Confidence Limits 148-178 114-160 92-141 80-122

107

108

Table 4.4. Mean survivorship and growth of 2-mo old juvenile Villosa iris exposed to different concentrations of mercuric chloride for 21-d.

* Significantly lower than control (p<0.05).

A= Mean in-water concentrations; out-water for all concentrations was below

detection limit (<8.4 µg/L).

B= Below detection limit (<8.4 µg/L) and expressed as nominal value.

Total [Hg]A (µg/L)

Survivorship (%)

Growth + SD (mm)

0B 90 0.51 + 0.17

4B 95 0.53 + 0.11

8B 100 0.37 + 0.13*

15 95 0.26 + 0.13*

32 100 0.06 + 0.05*

62 100 0.01 + 0.02*

114 100 0 + 0*

109

Figure 4.1 Test design of simulated environment; small scale microcosm used for chronic juvenile mussel toxicity tests.

AIR SUPPLY

110

Curriculum Vitae Theodore Walter Valenti

Home address: PO Box 373

Constableville, NY 13325 tvalenti@vt.edu

EDUCATION Graduate Master’s in Biology with emphasis in Environmental Toxicology 2004.

Virginia Polytechnic Institute and State University, Blacksburg VA. Primary Advisor: Dr. Donald S. Cherry. GPA: 3.6. Undergraduate Bachelor of Arts Degree 2001. Biology Major / Environmental Studies Minor.

Hamilton College, Clinton NY. Dean’s List Fall 2000, Spring 2001. Final GPA: 3.0. PROFESSIONAL INTERESTS/AFFILIATIONS

Ecotoxicology - Environmental risk assessment of freshwater systems through the examination of laboratory and field toxicological parameters.

- Integration of science and public affairs to create sound policy. - Development of standard test protocols.

Sigma Xi National Honor Society Society of Environmental Toxicology and Chemistry (SETAC) Save Our Streams Conservation Society (SOS) TEACHING EXPERIENCE Graduate Teaching Assistant. Virginia Tech, Department of Biology, Blacksburg VA.

Devised teaching schedules, lectured classes, administered examinations, and graded students’ work for introductory biology class (Spring 2002).

Substitute Teacher. South Lewis Central School District, Turin NY. Taught a variety of classes ranging in grade from K – 12. Also served as an assistant coach for varsity football and wrestling (Fall 2001).

Teaching Assistant. Hamilton College, Biology Department, Clinton NY

Prepared laboratories, assisted students, and conducted review sessions for introductory biology classes. (Fall 2000 – Spring 2001).

Teacher and Laboratory Assistant. Johns Hopkins’ Center for Talented Youth, Clinton NY.

Taught classes in Paleo-biology and Primate Evolution to gifted students ranging in age from 9 – 14 years (Summer 1999).

111

RESEARCH EXPERIENCE Graduate Research Assistant. Virginia Tech, Department of Biology, Blacksburg VA.

Conducted laboratory bioassays, carried out experiments in the field, and completely tasks pertinent to the daily operations of a toxicology laboratory (Fall 2002 – Fall 2004).

Research Assistant. New York State Department of Environmental Conservation, Rome NY.

Conducted laboratory bioassays on non-target aquatic organisms to assess the impact of pyrethroid insecticides currently applied to control the spread of the West Nile Virus (Fall 2000 - Spring 2001).

Research Assistant. Ernest Williams, Hamilton College, Biology Department, Clinton NY.

Observed organisms in the field, conducted behavioral studies, and examined host plant relationships (Summer 2000).

PUBLICATIONS Valenti TW, DS Cherry, RJ Neves, BA Locke, J Schmerfeld. 2004. Sensitivities of freshwater

mussel glochidia and regulatory test organisms to mercury and a reference toxicant. Invited chapter in SETAC book entitled “Bivalve Toxicology.” (In review).

Valenti TW, DS Cherry, RJ Neves, J Schmerfeld. 2004. Acute and chronic toxicity of mercury to

early life stages of the rainbow mussel, Villosa iris (Bivalvia: Unionidae)- Short Communication. Environ Toxic Contam 24 (in press).

Valenti TW, JL Chaffin, DS Cherry, ME. Schreiber, HM Valett, M Charles. 2004. Bioassessment

of an Appalachian headwater stream influenced by an abandoned arsenic mine. Arch Environ Toxicol Chem (In review).

Valenti TW, DS Cherry, RJ Neves, JW Jones, & RA Mair. Inter- and intra-specific

variability in the tolerance of glochidia exposed to sodium choride. (In prep).

Valenti TW, DS Cherry, RJ Neves. Acute and chronic exposure of early life stage of freshwater mussels to chlorine. (In prep).

GRANTS Cherry DS, RJ Currie, TW Valenti. An ecotoxicological evaluation of mercury in the North Fork Holston River at Saltville, Virginia. United States Fish & Wildlife Service ($58,000). Cherry DS, TW Valenti, RJ Currie, RJ Neves. First preliminary report on preassessment

screen data gaps Saltville national pollution list site natural resource damage assessment. United States Fish & Wildlife Service ($82,800).

Cherry DS, RJ Currie, TW Valenti, ML Simon, RJ Neves. Coal fines / particulate impact upon

mussel assemblages. United States Geological Survey ($32,000).

112

PRESENTATIONS / PUBLISHED ABSTRACTS Valenti TW, DS Cherry, RJ Neves, BA Locke. November 8-13th, 2003. Sensitivities of freshwater

mussel glochidia and regulatory test organisms to mercury and a reference toxicant. SETAC 24th Annual Meeting in North America. Austin, TX. Platform Presentation.

Valenti TW, DS Cherry, RJ Neves, BA Locke, JM Uerz, RJ Currie. November 8-13th, 2003.

Acute and chronic mercury exposures to different life stages of the rainbow mussel, Villosa iris. SETAC 24thAnnual Meeting in North America. Austin, TX. Poster Presentation.

Paul EA, HA Simonin, TW Valenti. November 16-20th, 2002. Toxicity of several mosquito

insecticides to crayfish. SETAC 23rd Annual Meeting in North America. Salt Lake City, UT. Poster Presentation.

Valenti TW, DH Catlin, EW Williams. July 28-30th, 2000. Some behavior of frosted elfins in lupine patches. The Annual Meeting of Lepidopterist’s Society, Wake Forest, Winston Salem, NC. Poster Presentation. Catlin DH, TW Valenti, EW Williams. November 3-4th, 2000. The effect of plant chemistry on oviposition of Phyciodes tharos butterflies. The 25th Annual Biological Sciences Research Symposium, Binghamton University, Binghamton, NY. Poster Presentation. TECHNICAL REPORTS Cherry DS, TW Valenti, RJ Currie, RJ Neves. First preliminary report on preassessment

screen data gaps Saltville national pollution list site natural resource damage assessment. (2003).

Cherry DS, TW Valenti, RJ Currie, RJ Neves, JW Jones. Chlorine toxicity to early life stages of freshwater mussels (2002-2004, quarterly).

Cherry DS, RJ Currie, TW Valenti, BA Locke, JM Uerz, ML Simon. Toxicity test laboratory

performance evaluation (DMR-QA 23 WET) for acute and/or chronic tests. (2003). Cherry DS, RJ Currie, TW Valenti, BA Locke, JM Uerz. Quarterly biomonitoring results of water flea and fathead minnow to Pine Bluff Mill Effluent, International Paper Co., Pine Bluff AR (2002-2003). Cherry DS, RJ Currie, TW Valenti, BA Locke, & JM Uerz. Acute / chronic toxicity evaluation of

Effluent 001 / 002 / 003 Celanese Acetate, Celco Plant, VA (2002-2004, quarterly).